INDEX

Explanations

lists of items or categories

oai_token-act-pair · gpt-3.5-turbo

phrases that include lists and enumerations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

matter

-0.79

imeter

-0.77

uca

-0.76

ibe

-0.74

imet

-0.73

umption

-0.72

jad

-0.72

ismo

-0.71

ester

-0.70

ixture

-0.69

POSITIVE LOGITS

 names

1.00

 accomplishments

0.97

 sorts

0.97

 grievances

0.97

 celebrities

0.89

 reasons

0.89

 items

0.88

 acron

0.87

 allowable

0.87

 possibilities

0.85

Activations Density 0.119%

lists of items or categories

phrases that include lists and enumerations

No Comments

No Known Activations