INDEX

Explanations

names and terms related to political personalities

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

nesday

-0.74

glers

-0.73

å§

-0.56

 Archdemon

-0.56

 Uganda

-0.55

 solvent

-0.55

 haunt

-0.52

behind

-0.51

bledon

-0.51

 Hots

-0.50

POSITIVE LOGITS

INAL

0.89

Ã£

0.83

oise

0.80

Spoiler

0.79

 Cortex

0.73

ĺħ

0.73

inosaur

0.69

ibrary

0.67

inal

0.67

ritic

0.66

Activations Density 1.613%

names and terms related to political personalities

No Comments

No Known Activations