INDEX

Explanations

references to reporters

oai_token-act-pair · gpt-3.5-turbo

mentions of reporters and journalism-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

emonic

-0.73

ensions

-0.71

 Causes

-0.71

BuyableInstoreAndOnline

-0.68

 dissolution

-0.66

 Defin

-0.66

uras

-0.66

phal

-0.65

 regeneration

-0.64

ujah

-0.64

POSITIVE LOGITS

 covering

1.04

 interviewed

0.89

 reporter

0.82

 journalists

0.81

orters

0.81

Journal

0.81

 reporters

0.80

 Seymour

0.80

 interviewing

0.80

 interrogated

0.77

Activations Density 0.088%

references to reporters

mentions of reporters and journalism-related terms

No Comments

No Known Activations