INDEX

Explanations

words and phrases related to involuntary actions or situations

oai_token-act-pair · gpt-3.5-turbo

terminologies related to involuntary actions or constraints

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 modelling

-0.73

Bay

-0.72

icated

-0.72

Sky

-0.71

Bou

-0.71

 optics

-0.69

 proven

-0.68

 refurb

-0.68

 Chand

-0.67

 hasht

-0.66

POSITIVE LOGITS

 involuntary

3.18

 invol

2.05

 forcibly

1.70

 forcefully

1.38

untary

1.35

untarily

1.35

 violently

1.34

Fug

1.31

ulp

1.26

 uncontroll

1.15

Activations Density 0.065%

words and phrases related to involuntary actions or situations

terminologies related to involuntary actions or constraints

No Comments

No Known Activations