INDEX

Explanations

phrases related to outcomes or results

oai_token-act-pair · gpt-3.5-turbo

phrases indicating potential outcomes or consequences

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

clips

-0.65

ashes

-0.62

vati

-0.61

verages

-0.57

ahead

-0.57

 discussed

-0.57

lund

-0.57

ida

-0.56

noticed

-0.55

 recounted

-0.55

POSITIVE LOGITS

be

1.49

Be

0.97

be

0.94

 contain

0.93

 resemble

0.90

asted

0.88

BE

0.86

 belong

0.86

 have

0.86

 consist

0.85

Activations Density 0.109%

phrases related to outcomes or results

phrases indicating potential outcomes or consequences

No Comments

No Known Activations