INDEX

Explanations

verbs associated with professional activities and processes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

dt

-0.77

elin

-0.71

none

-0.71

von

-0.67

txt

-0.66

html

-0.65

letter

-0.65

efer

-0.64

feat

-0.63

Ã¨re

-0.63

POSITIVE LOGITS

 assignments

0.84

 strangers

0.73

 oneself

0.70

 transitions

0.68

tains

0.68

 engagements

0.66

 drafts

0.66

 purchases

0.64

 purposes

0.64

 Directions

0.64

Activations Density 0.171%

verbs associated with professional activities and processes

No Comments

No Known Activations