INDEX

Explanations

reasons or explanations

oai_token-act-pair · gpt-3.5-turbo

explanations or justifications for statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

åĤ

-0.73

lem

-0.71

yan

-0.70

Winged

-0.70

SPONSORED

-0.69

shr

-0.68

scr

-0.68

âĹ¼

-0.67

Sham

-0.66

thro

-0.66

POSITIVE LOGITS

urers

0.91

endment

0.83

akening

0.82

rely

0.79

xual

0.75

pite

0.74

ecause

0.73

uristic

0.70

uesday

0.70

orus

0.70

Activations Density 0.051%

reasons or explanations

explanations or justifications for statements

No Comments

No Known Activations