INDEX

Explanations

references to authority figures or individuals in leadership roles

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

enegger

-0.69

ELL

-0.66

anus

-0.66

sung

-0.65

orph

-0.65

Sold

-0.64

æĺ

-0.63

gnu

-0.61

Boo

-0.61

milo

-0.60

POSITIVE LOGITS

 overseeing

1.06

boards

0.86

hyde

0.80

of

0.79

eering

0.73

 oversee

0.71

taker

0.69

Reviewer

0.68

 thereof

0.67

 helm

0.67

Activations Density 0.009%

references to authority figures or individuals in leadership roles

No Comments

No Known Activations