INDEX

Explanations

names of specific individuals

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_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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fecture

-0.79

worldly

-0.68

 lockout

-0.66

cipline

-0.66

 dracon

-0.65

 haunted

-0.64

ahime

-0.63

 unaccompanied

-0.63

prototype

-0.63

raints

-0.63

POSITIVE LOGITS

 Bernstein

0.98

 Shapiro

0.95

 Greenwald

0.95

 Huffington

0.89

 McInt

0.88

(@

0.86

 Tucker

0.84

News

0.84

 Friedman

0.83

 McCl

0.82

Activations Density 0.182%

names of specific individuals

No Comments

No Known Activations