INDEX

Explanations

terms associated with specific names, mentions, or titles

oai_token-act-pair · gpt-3.5-turbo

occurrences of specific nouns and verbs indicating people or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

iat

-0.88

Tat

-0.87

 Hound

-0.82

TI

-0.78

 Thib

-0.77

Sid

-0.77

iac

-0.77

NCT

-0.77

ript

-0.76

Iv

-0.76

POSITIVE LOGITS

 Greene

1.01

green

0.97

 Greenwald

0.93

Green

0.92

gre

0.91

 Morrison

0.89

ko

0.88

 Greenpeace

0.87

awa

0.87

 Green

0.87

Activations Density 0.399%

terms associated with specific names, mentions, or titles

occurrences of specific nouns and verbs indicating people or groups

No Comments

No Known Activations