INDEX

Explanations

instances of the word "hide" or phrases related to hiding or concealment

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ilingual

-0.75

iod

-0.71

bably

-0.67

etheless

-0.64

ipolar

-0.62

 Carnage

-0.62

odic

-0.61

 relegated

-0.60

theless

-0.59

isance

-0.58

POSITIVE LOGITS

away

1.11

ously

1.08

ous

0.94

 Caption

0.83

hiro

0.78

avi

0.75

aways

0.75

cases

0.73

hide

0.72

ance

0.72

Activations Density 0.005%

instances of the word "hide" or phrases related to hiding or concealment

No Comments

No Known Activations