INDEX

Explanations

issues with code syntax in a programming context

oai_token-act-pair · gpt-3.5-turbo

words indicating kindness or unkindness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

raged

-0.72

 victorious

-0.66

 executions

-0.66

otiation

-0.65

 marathon

-0.63

 sentencing

-0.62

 happier

-0.62

 recount

-0.62

 horrified

-0.60

 duel

-0.59

POSITIVE LOGITS

aterial

0.91

 ingredients

0.90

 intangible

0.86

 items

0.86

 components

0.85

 Element

0.85

hidden

0.83

 attributes

0.82

 Cosponsors

0.81

ãĥĥãĥī

0.79

Activations Density 1.009%

issues with code syntax in a programming context

words indicating kindness or unkindness

No Comments

No Known Activations