INDEX

Explanations

locations or landmarks such as plazas

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

%%%%

-0.68

erm

-0.67

ership

-0.60

lings

-0.58

lund

-0.57

loss

-0.57

quartered

-0.56

er

-0.56

hood

-0.56

legged

-0.54

POSITIVE LOGITS

xon

0.66

aurus

0.61

ptives

0.60

atl

0.59

Taj

0.59

 Seym

0.57

ption

0.57

 Plaza

0.55

ZI

0.55

arta

0.53

Activations Density 6.758%

locations or landmarks such as plazas

No Comments

No Known Activations