INDEX

Explanations

words or phrases that imply significant consequences or impact

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

 Chattanooga

-0.73

 scram

-0.71

 Sammy

-0.65

 scrambled

-0.64

 Hilton

-0.62

 Reverend

-0.62

 Mobil

-0.62

 fisherman

-0.61

 Dillon

-0.60

jog

-0.60

POSITIVE LOGITS

ï¸ı

1.07

_>

0.98

STEM

0.96

whether

0.92

âĶĢâĶĢâĶĢâĶĢ

0.92

minus

0.91

actual

0.88

Balt

0.87

where

0.82

physical

0.82

Activations Density 0.338%

words or phrases that imply significant consequences or impact

No Comments

No Known Activations