INDEX

Explanations

phrases related to causes or reasons

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hops

-0.65

itutes

-0.63

ogun

-0.62

eds

-0.62

ours

-0.62

tops

-0.61

agues

-0.61

etc

-0.60

 harms

-0.59

eks

-0.59

POSITIVE LOGITS

 inability

0.89

 insistence

0.86

 willingness

0.82

 consistency

0.79

 unwillingness

0.76

xual

0.75

 realization

0.75

 acknowledgement

0.73

 understatement

0.72

BILITY

0.71

Activations Density 0.377%

phrases related to causes or reasons

No Comments

No Known Activations