INDEX

Explanations

references to people and their relationships, often highlighting inclusion or discussion of community

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

advertisement

-0.79

naires

-0.73

 Scully

-0.72

classic

-0.66

kamp

-0.65

 Daniels

-0.64

odor

-0.63

rique

-0.62

ĵ

-0.61

 ..............

-0.61

POSITIVE LOGITS

'd

1.03

 resided

0.96

 belonged

0.92

 reside

0.91

 geographically

0.91

 belong

0.88

 stayed

0.84

're

0.83

 lived

0.83

 nearest

0.82

Activations Density 0.055%

references to people and their relationships, often highlighting inclusion or discussion of community

No Comments

No Known Activations