INDEX

Explanations

mentions of or references to reporters

oai_token-act-pair · gpt-3.5-turbo

mentions of reporters in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inence

-0.77

perm

-0.73

ministic

-0.72

ertodd

-0.72

ensions

-0.72

ateral

-0.72

aline

-0.68

emonic

-0.68

icted

-0.68

uras

-0.68

POSITIVE LOGITS

cov

0.88

 reporter

0.84

 reporters

0.83

verage

0.78

 covering

0.73

 coverage

0.72

 corps

0.71

 Coverage

0.70

 COVER

0.70

 journalists

0.70

Activations Density 0.033%

mentions of or references to reporters

mentions of reporters in various contexts

No Comments

No Known Activations