INDEX

Explanations

people or sources mentioned in news articles

oai_token-act-pair · gpt-3.5-turbo

phrases indicating sources or references to individuals knowledgeable about a situation

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

Interstitial

-0.75

-0.74

ãĥĺ

-0.73

 landfall

-0.67

tumblr

-0.66

osate

-0.65

 prayers

-0.65

ennial

-0.64

 unconstitutional

-0.64

-0.63

POSITIVE LOGITS

 briefed

1.08

 knowledgeable

0.95

 familiar

0.89

 insider

0.82

 priv

0.81

 confid

0.80

 disclosed

0.78

amon

0.75

remlin

0.74

ispers

0.74

Activations Density 0.302%

people or sources mentioned in news articles

phrases indicating sources or references to individuals knowledgeable about a situation

No Comments

No Known Activations