INDEX

Explanations

references to influential individuals or significant public figures

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

okemon

-0.71

 Ples

-0.67

Operation

-0.67

ilation

-0.67

oise

-0.65

prus

-0.63

 contrace

-0.62

mble

-0.62

IFE

-0.61

 Congo

-0.61

POSITIVE LOGITS

head

1.16

heads

1.02

 prominently

0.97

 skating

0.89

aces

0.86

enance

0.82

aced

0.77

heading

0.76

polit

0.75

hood

0.74

Activations Density 0.012%

references to influential individuals or significant public figures

No Comments

No Known Activations