INDEX

Explanations

words associated with deception and isolation

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

PDATE

-0.74

theless

-0.73

 Shack

-0.71

 Dragonbound

-0.69

 Doctrine

-0.66

 Belt

-0.64

 miscarriage

-0.64

 Penet

-0.63

 LORD

-0.62

 Heller

-0.61

POSITIVE LOGITS

ations

1.94

ating

1.81

ates

1.73

ators

1.70

atory

1.56

ator

1.55

ational

1.51

ative

1.43

ated

1.43

ate

1.34

Activations Density 0.025%

words associated with deception and isolation

No Comments

No Known Activations