INDEX

Explanations

mentions of "effects" related to various subjects or phenomena

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

swick

-0.83

mint

-0.75

apest

-0.74

croft

-0.65

 mant

-0.64

ramid

-0.63

mbuds

-0.63

ERO

-0.60

theless

-0.60

ardless

-0.60

POSITIVE LOGITS

effects

1.15

iveness

1.09

 effects

0.90

 exerted

0.88

bringer

0.84

 thereof

0.83

 outweigh

0.83

 wrought

0.81

Effect

0.81

olation

0.78

Activations Density 0.015%

mentions of "effects" related to various subjects or phenomena

No Comments

No Known Activations