INDEX

Explanations

comparisons or emphasis of a degree in text, especially related to evaluating effectiveness or significance

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

lict

-0.69

ilings

-0.65

iners

-0.65

encer

-0.63

 livest

-0.63

uster

-0.61

iasco

-0.60

 prosec

-0.58

ividual

-0.58

 Prosecut

-0.58

POSITIVE LOGITS

fare

1.08

ward

0.96

point

0.92

points

0.90

finding

0.89

forward

0.81

WARD

0.80

YY

0.77

liness

0.77

bos

0.76

Activations Density 11.543%

comparisons or emphasis of a degree in text, especially related to evaluating effectiveness or significance

No Comments

No Known Activations