INDEX

Explanations

references to specific individuals, particularly public figures and their actions

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

points

-0.77

binary

-0.69

cape

-0.69

isted

-0.69

 FranÃ§

-0.68

stice

-0.68

istance

-0.68

opausal

-0.67

ãĥĦ

-0.67

itarian

-0.67

POSITIVE LOGITS

 McCabe

0.96

hew

0.86

hews

0.73

atcher

0.73

plot

0.71

onduct

0.70

shaw

0.67

ursed

0.66

0.65

abouts

0.64

Activations Density 0.005%

references to specific individuals, particularly public figures and their actions

No Comments

No Known Activations