INDEX

Explanations

phrases related to expert opinions or statements in text

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

\\\\\\\\

-0.69

âĸ¬âĸ¬

-0.66

 Rabbit

-0.61

LET

-0.61

ãĥ¼ãĥĨãĤ£

-0.61

Barn

-0.60

ictional

-0.59

 Scrib

-0.59

Letter

-0.57

ptives

-0.57

POSITIVE LOGITS

hops

0.68

yk

0.68

 challeng

0.66

mson

0.64

 coordin

0.61

ATK

0.61

hement

0.61

 adding

0.59

ca

0.59

enz

0.59

Activations Density 7.971%

phrases related to expert opinions or statements in text

No Comments

No Known Activations