INDEX

Explanations

references to authorship or attribution in the context of photography or media

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

conservancy

-0.80

amphetamine

-0.77

TPP

-0.73

cffff

-0.73

competitive

-0.72

Contents

-0.69

versions

-0.69

terms

-0.69

balanced

-0.68

entimes

-0.67

POSITIVE LOGITS

 Ibid

1.13

 Dave

0.94

 Anonymous

0.93

 Ezra

0.92

Ian

0.90

 Cynthia

0.89

 Sarah

0.89

 Evan

0.88

 Mark

0.88

 Jacqu

0.87

Activations Density 0.046%

references to authorship or attribution in the context of photography or media

No Comments

No Known Activations