INDEX

Explanations

transcripts

oai_token-act-pair · gpt-3.5-turbo

references to transcripts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

vil

-0.72

ggle

-0.71

yu

-0.70

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

-0.66

zzle

-0.65

deen

-0.64

pper

-0.63

 Bengal

-0.62

dream

-0.62

 Fury

-0.62

POSITIVE LOGITS

 Transcript

1.26

 transcripts

1.26

 transcript

1.21

 transcription

1.00

icons

0.84

 snippets

0.82

ions

0.79

ophone

0.78

ophon

0.78

 excerpts

0.76

Activations Density 0.013%

transcripts

references to transcripts

No Comments

No Known Activations