INDEX

Explanations

references to xenophobia or related terms

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

 DRAG

-0.76

MENTS

-0.74

 captcha

-0.73

LOAD

-0.72

theless

-0.72

IRD

-0.72

 LINE

-0.71

LEASE

-0.70

HEAD

-0.69

UGH

-0.69

POSITIVE LOGITS

ophobic

1.60

ophobia

1.44

ophon

1.28

ophob

1.25

ocide

1.25

omorph

1.17

obia

1.09

etics

1.04

opolis

1.02

obl

1.00

Activations Density 0.002%

references to xenophobia or related terms

No Comments

No Known Activations