INDEX

Explanations

mentions of social interactions and casual conversations

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

 ÂŃ

-0.91

 strikingly

-0.71

Footnote

-0.70

—

-0.69

âĢĳ

-0.68

Newsletter

-0.65

 Lyndon

-0.64

 extraordinarily

-0.64

 Rodham

-0.64

 policymakers

-0.62

POSITIVE LOGITS

 doesnt

1.34

;)

1.30

lol

1.30

 didnt

1.27

 haha

1.25

 dont

1.22

 alot

1.18

tho

1.14

:)

1.12

 kinda

1.10

Activations Density 0.659%

mentions of social interactions and casual conversations

No Comments

No Known Activations