INDEX

Explanations

entities associated with music industry accolades and music-related terms like 'lyrical' and 'caliber'

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wal

-0.61

patrick

-0.59

walk

-0.59

hire

-0.58

pring

-0.57

cloth

-0.56

busters

-0.56

eur

-0.56

hari

-0.56

liness

-0.56

POSITIVE LOGITS

XM

0.72

 tremend

0.67

emonic

0.60

iscopal

0.59

ICAN

0.57

senal

0.56

uble

0.52

ONSORED

0.52

 chloride

0.51

itone

0.51

Activations Density 7.249%

entities associated with music industry accolades and music-related terms like 'lyrical' and 'caliber'

No Comments

No Known Activations