INDEX

Explanations

instances of the word "to."

oai_token-act-pair · gpt-4o-mini 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

irection

-0.71

CHAT

-0.67

Developer

-0.64

awn

-0.62

IPP

-0.61

UTH

-0.59

MpServer

-0.59

rowth

-0.57

Ê

-0.57

ace

-0.56

POSITIVE LOGITS

 appease

0.89

 repertoire

0.88

 bolster

0.86

 compensate

0.85

 complement

0.85

mix

0.81

 reperto

0.78

 satisfy

0.77

orate

0.72

 list

0.72

Activations Density 0.089%

instances of the word "to."

No Comments

No Known Activations