INDEX

Explanations

movie titles or keywords related to media and entertainment

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lement

-0.80

rals

-0.74

aram

-0.73

ral

-0.73

ahime

-0.71

mental

-0.70

ment

-0.70

riel

-0.70

orney

-0.67

mine

-0.66

POSITIVE LOGITS

Cola

0.85

 Bros

0.74

Books

0.73

Arcade

0.71

Pocket

0.70

 Wrap

0.69

 Comics

0.66

Fest

0.65

 Wiki

0.64

 Girls

0.63

Activations Density 0.160%

movie titles or keywords related to media and entertainment

No Comments

No Known Activations