INDEX

Explanations

items and actions related to a fantasy game setting, potentially involving combat and spells

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

pmwiki

-0.65

ername

-0.61

guyen

-0.60

ãĥķãĤ¡

-0.60

 veter

-0.58

 GOODMAN

-0.57

ãĥ¼ãĥĨãĤ£

-0.57

combe

-0.56

itton

-0.55

ãĥ¼ãĥĨ

-0.54

POSITIVE LOGITS

 elephants

0.66

 demons

0.65

 emperor

0.65

aged

0.60

 demonic

0.60

aging

0.59

 blink

0.58

 wolves

0.57

 hydra

0.56

 pandemonium

0.56

Activations Density 12.501%

items and actions related to a fantasy game setting, potentially involving combat and spells

No Comments

No Known Activations