INDEX

Explanations

common terms or phrases that signal a transitioning or shifting of topics or thoughts within a larger piece of writing

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

schild

-0.73

livious

-0.73

raid

-0.71

dating

-0.70

codes

-0.68

mast

-0.66

zza

-0.64

zo

-0.64

 Seym

-0.63

zan

-0.63

POSITIVE LOGITS

 forth

1.12

endum

0.81

 together

0.77

 forward

0.74

up

0.72

 attention

0.72

unity

0.71

 smiles

0.67

 misfortune

0.67

ageddon

0.66

Activations Density 7.202%

common terms or phrases that signal a transitioning or shifting of topics or thoughts within a larger piece of writing

No Comments

No Known Activations