INDEX

Explanations

words related to investigations or searches

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

 GOODMAN

-0.73

 whales

-0.64

GGGG

-0.61

 underwater

-0.61

 disapp

-0.58

 Predator

-0.58

 Banks

-0.57

 UFOs

-0.56

ay

-0.55

++++++++

-0.54

POSITIVE LOGITS

kov

0.85

esy

0.84

uble

0.84

enegger

0.78

orst

0.77

itzer

0.77

lich

0.76

ifer

0.76

wagen

0.75

vere

0.75

Activations Density 0.039%

words related to investigations or searches

No Comments

No Known Activations