INDEX

Explanations

terms related to comparisons and analogies

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_slefr-ajt/2-res_slefr-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.2.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

tsy

-0.95

ding

-0.90

dot

-0.89

ieri

-0.88

 Epidem

-0.87

bed

-0.86

adia

-0.84

sburg

-0.83

tab

-0.80

tailed

-0.80

POSITIVE LOGITS

iour

1.02

 comparison

0.97

isons

0.92

 warr

0.91

 comparisons

0.91

 Compar

0.91

 compare

0.90

 Compare

0.90

 compares

0.89

Osw

0.89

Activations Density 0.166%

terms related to comparisons and analogies

No Comments

No Known Activations