INDEX

Explanations

words or phrases related to norms, standards, or common occurrences

oai_token-act-pair · gpt-3.5-turbo

phrases indicating general trends or typical behaviors

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

htaking

-0.70

atures

-0.69

Reviewer

-0.68

á

-0.66

uel

-0.66

udi

-0.63

ati

-0.63

ancer

-0.63

arta

-0.62

assi

-0.62

POSITIVE LOGITS

entimes

1.16

 accompanied

0.90

 consist

0.90

 abbrevi

0.85

 involve

0.85

 preceded

0.82

 consists

0.82

 overlooked

0.82

 followed

0.81

 referred

0.81

Activations Density 0.110%

words or phrases related to norms, standards, or common occurrences

phrases indicating general trends or typical behaviors

No Comments

No Known Activations