INDEX

Explanations

financial terms and terms related to analysis

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

ĸļ

-0.74

boats

-0.60

 Marathon

-0.58

DeL

-0.56

pty

-0.55

 Mamm

-0.54

boat

-0.54

anski

-0.53

 raft

-0.53

frey

-0.53

POSITIVE LOGITS

 analyst

0.89

alyst

0.75

 analysts

0.70

 Analyst

0.69

irtual

0.62

swick

0.61

ocular

0.59

igible

0.58

ariat

0.58

 speculate

0.57

Activations Density 6.614%

financial terms and terms related to analysis

No Comments

No Known Activations