INDEX

Explanations

references to events, organizations, and places

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the."

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

 besides

-0.75

pots

-0.73

 teasp

-0.72

pai

-0.70

olt

-0.70

olen

-0.69

raint

-0.68

pler

-0.68

zac

-0.66

being

-0.66

POSITIVE LOGITS

 Supreme

0.95

 Associated

0.90

 Subcommittee

0.89

 Department

0.89

 Securities

0.88

 Board

0.88

 venerable

0.87

 ACLU

0.85

 Bureau

0.84

 Institute

0.83

Activations Density 0.221%

references to events, organizations, and places

occurrences of the word "the."

No Comments

No Known Activations