INDEX

Explanations

mentions of news agencies or media outlets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

starter

-0.70

adobe

-0.70

atorium

-0.69

Ì¶

-0.65

starting

-0.65

powers

-0.64

termin

-0.63

ighton

-0.63

 Bowser

-0.63

cus

-0.62

POSITIVE LOGITS

 VIDEOS

1.00

EG

0.85

HY

0.84

NEWS

0.84

ews

0.83

 PRESS

0.80

INESS

0.78

 IMAGES

0.78

olitics

0.76

 News

0.76

Activations Density 0.020%

mentions of news agencies or media outlets

No Comments

No Known Activations