INDEX

Explanations

names, particularly with the pattern "-sen" at the end

oai_token-act-pair · gpt-3.5-turbo

the word "sen" followed by any other token

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

Pwr

-0.73

 tolerance

-0.73

 hitch

-0.65

 heels

-0.64

tails

-0.63

 casing

-0.63

 chromosome

-0.63

 Adin

-0.62

 grounding

-0.62

 anticipation

-0.62

POSITIVE LOGITS

pai

1.59

iors

1.13

ior

1.09

iture

1.05

ault

1.00

vironment

0.99

egal

0.98

seless

0.97

ocide

0.96

escent

0.96

Activations Density 0.009%

names, particularly with the pattern "-sen" at the end

the word "sen" followed by any other token

No Comments

No Known Activations