INDEX

Explanations

proper nouns and names, potentially related to technology, companies, or individuals

oai_token-act-pair · gpt-3.5-turbo

references to intellectual properties or media formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 moreover

-0.62

 Spot

-0.62

than

-0.60

dated

-0.59

Production

-0.58

immer

-0.58

agraph

-0.57

depending

-0.55

Chain

-0.54

 furthermore

-0.54

POSITIVE LOGITS

did

1.08

 does

0.86

did

0.84

do

0.82

 predecessors

0.81

 counterparts

0.75

 predicted

0.74

 envisioned

0.73

 hoped

0.71

 would

0.65

Activations Density 0.219%

proper nouns and names, potentially related to technology, companies, or individuals

references to intellectual properties or media formats

No Comments

No Known Activations