INDEX

Explanations

phrases related to legal and political contexts, such as subpoenas, investigations, and political actions

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

theless

-0.61

joy

-0.54

 regards

-0.53

packing

-0.52

 regard

-0.49

ning

-0.48

ishly

-0.48

draft

-0.48

ishing

-0.48

ista

-0.48

POSITIVE LOGITS

atics

0.95

atically

0.84

eworks

0.75

ateg

0.71

etrical

0.71

yth

0.70

etr

0.69

acists

0.69

acies

0.68

 oper

0.67

Activations Density 11.935%

phrases related to legal and political contexts, such as subpoenas, investigations, and political actions

No Comments

No Known Activations