INDEX

Explanations

instances of the word "stood to", usually followed by a verb indicating the potential outcome or expectation of a situation

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "to" with various contexts implying intentions or potential outcomes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

efficients

-0.63

hops

-0.61

 elig

-0.61

emort

-0.61

accompanied

-0.58

gallery

-0.58

packing

-0.57

 referral

-0.57

trip

-0.57

 jails

-0.56

POSITIVE LOGITS

ggles

0.97

asty

0.93

asted

0.92

asts

0.89

oth

0.81

othy

0.79

asting

0.78

pless

0.77

wing

0.73

 preserve

0.72

Activations Density 0.239%

instances of the word "stood to", usually followed by a verb indicating the potential outcome or expectation of a situation

occurrences of the phrase "to" with various contexts implying intentions or potential outcomes

No Comments

No Known Activations