INDEX

Explanations

reporting on political actions and statements

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

olars

-0.77

ept

-0.73

aic

-0.72

zens

-0.70

byss

-0.69

iatures

-0.68

"}],"

-0.67

ateral

-0.67

haar

-0.67

pite

-0.66

POSITIVE LOGITS

 himself

1.00

 aloud

0.93

 Mexicans

0.87

 openly

0.86

 famously

0.82

 homosexuals

0.81

 insulting

0.80

 onstage

0.80

 gays

0.80

 controversial

0.79

Activations Density 0.232%

reporting on political actions and statements

No Comments

No Known Activations