INDEX

Explanations

phrases related to arguments against certain claims or actions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Presence

-0.80

iership

-0.80

ideon

-0.71

assies

-0.71

ahime

-0.70

lycer

-0.68

RAFT

-0.66

alde

-0.64

 Suite

-0.63

 Prix

-0.62

POSITIVE LOGITS

xious

1.18

except

0.99

oses

0.95

ct

0.94

 avail

0.94

vell

0.92

 longer

0.92

 matter

0.89

xus

0.89

obs

0.89

Activations Density 4.334%

phrases related to arguments against certain claims or actions

No Comments

No Known Activations

phrases related to arguments against certain claims or actions

No Comments

No Known Activations