INDEX

Explanations

instances of the word "pay" and related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to payment and salary

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

GGGGGGGG

-0.80

âĸ¬

-0.78

 Scotia

-0.75

ãĤ¼ãĤ¦ãĤ¹

-0.74

FTWARE

-0.72

ãĥķ

-0.71

EStream

-0.70

_>

-0.70

ARGET

-0.69

Jub

-0.68

POSITIVE LOGITS

checks

1.29

outs

1.20

offs

1.02

roll

0.97

paying

0.92

paid

0.90

 paid

0.89

rites

0.89

pal

0.88

 dues

0.87

Activations Density 0.036%

instances of the word "pay" and related terms

terms related to payment and salary

No Comments

No Known Activations