INDEX

Explanations

names of authors or contributors in written texts

oai_token-act-pair · gpt-3.5-turbo

references to authors or contributors in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 prey

-0.72

 favourites

-0.71

 predators

-0.71

 bystanders

-0.70

 queues

-0.68

 turnover

-0.67

 offenders

-0.66

qv

-0.65

 reward

-0.65

 vigilant

-0.64

POSITIVE LOGITS

Edited

0.92

Synopsis

0.89

Iss

0.89

Editor

0.88

WASHINGTON

0.84

Nation

0.84

Dear

0.82

Narr

0.82

Ill

0.81

EVA

0.81

Activations Density 0.113%

names of authors or contributors in written texts

references to authors or contributors in a document

No Comments

No Known Activations