INDEX

Explanations

quotes from or mentions of reporters

oai_token-act-pair · gpt-3.5-turbo

mentions of reporters and journalism-related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ston

-0.86

vasive

-0.69

ateral

-0.66

erton

-0.63

uras

-0.62

spring

-0.62

fin

-0.62

emonic

-0.61

fee

-0.60

qual

-0.60

POSITIVE LOGITS

 reporters

0.92

cov

0.82

 reporter

0.81

hip

0.76

 journalists

0.76

orters

0.75

 photographers

0.74

etter

0.73

 Journalists

0.73

hips

0.73

Activations Density 0.017%

quotes from or mentions of reporters

mentions of reporters and journalism-related contexts

No Comments

No Known Activations