INDEX

Explanations

instances of the word "cast" and its variations, indicating a focus on actions or abilities related to casting in a gaming or magical context

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

iliate

-0.57

kef

-0.56

 Haram

-0.55

 Staples

-0.53

 jams

-0.53

hus

-0.52

 Harding

-0.51

hook

-0.51

alus

-0.51

cester

-0.51

POSITIVE LOGITS

 cast

0.56

 precision

0.55

 healed

0.53

 breadth

0.50

pelling

0.49

 radius

0.47

 backgrounds

0.47

 wonder

0.46

 leveled

0.46

 doubt

0.45

Activations Density 0.001%

instances of the word "cast" and its variations, indicating a focus on actions or abilities related to casting in a gaming or magical context

No Comments

No Known Activations