INDEX

Explanations

terms related to exploring or investigating a variety of topics or objects

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

ivari

-0.61

jud

-0.61

fixed

-0.58

iah

-0.58

fight

-0.58

lat

-0.55

gage

-0.53

Ĩ

-0.52

processing

-0.52

activate

-0.51

POSITIVE LOGITS

ationally

0.79

vier

0.78

nels

0.71

ibility

0.71

ibilities

0.69

schild

0.68

 avenues

0.68

 possibilities

0.68

Ô

0.67

ĸļ

0.66

Activations Density 13.776%

terms related to exploring or investigating a variety of topics or objects

No Comments

No Known Activations