INDEX

Explanations

phrases referring to ways in which actions or scenarios are compared or related

oai_token-act-pair · gpt-3.5-turbo

comparisons that express similarity or analogy between different subjects or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

igh

-0.73

mun

-0.62

ONSORED

-0.60

Throw

-0.59

eri

-0.58

ategor

-0.57

 McGee

-0.56

bart

-0.56

 compe

-0.55

throw

-0.55

POSITIVE LOGITS

ettings

0.79

 rapists

0.70

ounter

0.70

isSpecialOrderable

0.69

achu

0.69

abl

0.68

liness

0.65

ractor

0.64

aws

0.61

 Cooke

0.61

Activations Density 0.045%

phrases referring to ways in which actions or scenarios are compared or related

comparisons that express similarity or analogy between different subjects or concepts

No Comments

No Known Activations