INDEX

Explanations

phrases related to ensuring something is done correctly or effectively

oai_token-act-pair · gpt-3.5-turbo

instances of the word "properly" in relation to processes or actions being executed correctly

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ista

-0.85

oleon

-0.83

iens

-0.83

ĸļ

-0.79

aily

-0.72

İ

-0.72

raltar

-0.71

isms

-0.71

————————

-0.69

ists

-0.68

POSITIVE LOGITS

 behaved

0.94

 positioned

0.94

 aligned

0.92

 formatted

0.88

 handled

0.86

 communicated

0.84

 diagnosed

0.83

 understood

0.83

 dispose

0.82

 clothed

0.82

Activations Density 0.041%

phrases related to ensuring something is done correctly or effectively

instances of the word "properly" in relation to processes or actions being executed correctly

No Comments

No Known Activations