INDEX

Explanations

phrases or words related to proof or evidence

oai_token-act-pair · gpt-3.5-turbo

references to proof or validation of claims or assertions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 livest

-0.84

ecycle

-0.71

 contrace

-0.67

 Peninsula

-0.65

cffffcc

-0.64

kson

-0.64

ideshow

-0.64

pta

-0.63

tyr

-0.63

 newsletters

-0.63

POSITIVE LOGITS

reading

1.35

reader

1.22

read

1.14

 proofs

0.85

proof

0.82

 proof

0.78

positive

0.77

ificate

0.77

uers

0.77

 pudding

0.77

Activations Density 0.025%

phrases or words related to proof or evidence

references to proof or validation of claims or assertions

No Comments

No Known Activations