INDEX

Explanations

phrases prompting action or response from the reader

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

peak

-0.69

etting

-0.65

apo

-0.65

stration

-0.62

uit

-0.62

etter

-0.61

rift

-0.59

raft

-0.57

road

-0.56

definition

-0.56

POSITIVE LOGITS

 been

1.13

gotten

1.04

 Been

1.01

 gotten

0.97

been

0.95

 begun

0.85

zens

0.83

 grown

0.83

 undergone

0.82

 gone

0.79

Activations Density 0.076%

phrases prompting action or response from the reader

No Comments

No Known Activations