INDEX

Explanations

instances of the word "go."

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

phrases related to movement or transition

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

uctor

-0.77

creen

-0.76

itionally

-0.70

 Horus

-0.66

ament

-0.65

ricted

-0.64

ificent

-0.63

dimension

-0.61

ullah

-0.59

oret

-0.59

POSITIVE LOGITS

lems

1.01

vt

0.91

verning

0.91

 Forth

0.87

ogly

0.85

ogl

0.84

ggle

0.84

 forth

0.76

 limp

0.72

GO

0.72

Activations Density 0.047%

instances of the word "go."

phrases related to movement or transition

No Comments

No Known Activations