INDEX

Explanations

the word "no" followed by a numerical value indicating the strength of negation

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a lack of or absence of something

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

lass

-0.72

rex

-0.69

DragonMagazine

-0.65

late

-0.64

POST

-0.62

srf

-0.62

merce

-0.62

often

-0.61

println

-0.61

RAFT

-0.61

POSITIVE LOGITS

 doubt

1.04

 earthly

1.01

 clue

0.96

 longer

0.92

 recourse

0.89

 qual

0.88

 conceivable

0.87

xious

0.87

 chance

0.87

 shortage

0.86

Activations Density 0.044%

the word "no" followed by a numerical value indicating the strength of negation

phrases indicating a lack of or absence of something

No Comments

No Known Activations