INDEX

Explanations

proper nouns, likely related to news articles or headlines

oai_token-act-pair · gpt-3.5-turbo

references to events or statements related to political contexts

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

worms

-0.62

 ILCS

-0.62

,,,,

-0.60

 organisers

-0.55

Advertisements

-0.55

adobe

-0.55

IPM

-0.55

seller

-0.53

 pigeon

-0.53

 Centauri

-0.52

POSITIVE LOGITS

 POLITICO

0.90

Reuters

0.79

UTERS

0.77

 Priebus

0.75

 Bannon

0.75

 Politico

0.72

 Reuters

0.71

AP

0.71

 Getty

0.71

AP

0.68

Activations Density 0.188%

proper nouns, likely related to news articles or headlines

references to events or statements related to political contexts

No Comments

No Known Activations