INDEX

Explanations

mentions of political involvement, investigations, and possibly criminal activities

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bern

-0.50

geons

-0.50

amina

-0.49

hest

-0.47

Scient

-0.47

oiler

-0.46

sell

-0.45

geries

-0.45

Blocks

-0.45

ãĥ£

-0.44

POSITIVE LOGITS

ateral

0.49

naire

0.49

uality

0.47

ISM

0.46

yip

0.45

worthiness

0.45

orship

0.44

 destro

0.44

ality

0.44

 HELL

0.44

Activations Density 6.934%

mentions of political involvement, investigations, and possibly criminal activities

No Comments

No Known Activations