INDEX

Explanations

content related to news articles or statements from official sources

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

 improv

-0.73

 stump

-0.72

entimes

-0.70

 temptation

-0.68

 jugg

-0.67

 crou

-0.66

 charisma

-0.66

 trivia

-0.66

 experien

-0.65

 fascination

-0.64

POSITIVE LOGITS

Statement

0.92

emphasis

0.90

itled

0.86

 titled

0.81

lishes

0.81

 redacted

0.80

idelines

0.79

 stated

0.79

 "...

0.79

 letter

0.79

Activations Density 1.740%

content related to news articles or statements from official sources

No Comments

No Known Activations