INDEX

Explanations

proper nouns related to people

oai_token-act-pair · gpt-3.5-turbo

references to individuals in a context

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

 Bound

-0.70

ulp

-0.69

 economical

-0.63

Georg

-0.62

ÂłÂłÂłÂł

-0.59

 Guys

-0.59

Nec

-0.59

ELF

-0.59

AGE

-0.58

interstitial

-0.57

POSITIVE LOGITS

soever

0.94

ophon

0.94

 frequ

0.88

 accompanies

0.87

 redes

0.86

oping

0.85

 oversaw

0.85

 livest

0.84

 resided

0.84

 oversees

0.83

Activations Density 0.097%

proper nouns related to people

references to individuals in a context

No Comments

No Known Activations