INDEX

Explanations

references to religious figures, specifically those related to the title "Reverend."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

holders

-0.83

bags

-0.69

boxes

-0.65

remlin

-0.64

falls

-0.63

 proximity

-0.61

 stuffing

-0.60

 explosives

-0.60

WAYS

-0.59

aud

-0.59

POSITIVE LOGITS

isions

1.28

olutions

1.20

olt

1.06

olver

1.02

ayne

1.01

ivals

1.00

olving

0.98

ision

0.95

ocation

0.93

olves

0.93

Activations Density 0.016%

references to religious figures, specifically those related to the title "Reverend."

No Comments

No Known Activations