INDEX

Explanations

technical terms denoting specific definitions or instructions

oai_token-act-pair · gpt-3.5-turbo

terms and phrases related to definitions and default characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 cris

-0.65

wait

-0.62

 Vinyl

-0.61

raped

-0.61

 prim

-0.60

psy

-0.58

ufact

-0.57

 Pupp

-0.57

 Shine

-0.57

 belie

-0.57

POSITIVE LOGITS

 guiActiveUn

0.85

tera

0.76

tesy

0.73

idth

0.71

elson

0.71

ĳå£«

0.68

eus

0.68

wik

0.66

classes

0.66

loading

0.65

Activations Density 0.050%

technical terms denoting specific definitions or instructions

terms and phrases related to definitions and default characteristics

No Comments

No Known Activations