INDEX

Explanations

phrases related to public information, sharing or legal actions

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

Scroll

-0.76

imov

-0.71

jack

-0.68

Reloaded

-0.67

pread

-0.67

ikawa

-0.66

Smith

-0.65

RANT

-0.65

pell

-0.64

scrib

-0.64

POSITIVE LOGITS

 outcry

0.89

 perception

0.88

 servants

0.87

eye

0.86

 purse

0.82

 sector

0.81

 broadcaster

0.80

 sphere

0.80

 opinion

0.78

 relations

0.78

Activations Density 0.039%

phrases related to public information, sharing or legal actions

No Comments

No Known Activations