INDEX

Explanations

phrases emphasizing inclusion or universality

oai_token-act-pair · gpt-3.5-turbo

repetitive conjunctions indicating inclusivity or expansiveness

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

 Corpus

-0.60

 Authors

-0.59

Beg

-0.58

pec

-0.56

Pick

-0.56

pi

-0.53

 Crunch

-0.53

Wire

-0.52

 detectors

-0.52

Mess

-0.52

POSITIVE LOGITS

rogen

1.09

 sund

0.93

rogens

0.91

ro

0.91

 every

0.87

 EVERY

0.84

 everyone

0.80

every

0.80

ãĥ¼ãĥĨ

0.79

 everything

0.77

Activations Density 0.049%

phrases emphasizing inclusion or universality

repetitive conjunctions indicating inclusivity or expansiveness

No Comments

No Known Activations