INDEX

Explanations

references to specific actions or behaviors in a narrative context

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

ĪĴ

-0.74

 variance

-0.71

dayName

-0.71

 showc

-0.70

omaly

-0.68

 tremend

-0.67

ãĥ¯ãĥ³

-0.65

llah

-0.65

 lapt

-0.65

Ø¹

-0.62

POSITIVE LOGITS

enges

0.97

doms

0.83

enge

0.80

enstein

0.76

angles

0.76

eus

0.74

owship

0.74

ocket

0.74

enthal

0.71

Ã¨re

0.71

Activations Density 0.016%

references to specific actions or behaviors in a narrative context

No Comments

No Known Activations