INDEX

Explanations

references to significant illumination or highlighting within events or contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_slefr-ajt/2-res_slefr-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.2.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

 partnerships

-0.67

 partnership

-0.64

iva

-0.62

hire

-0.62

pair

-0.60

Tex

-0.58

acio

-0.58

asm

-0.57

 contract

-0.56

bound

-0.56

POSITIVE LOGITS

 Reef

0.71

 Niet

0.68

dw

0.67

 Pavilion

0.67

Dah

0.66

 Yuan

0.66

allery

0.63

 Glac

0.62

heid

0.62

summary

0.61

Activations Density 0.000%

references to significant illumination or highlighting within events or contexts

No Comments

No Known Activations