INDEX

Explanations

references to mothers and parenting-related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Flavoring

-1.06

 Heist

-0.86

aneously

-0.77

OLOG

-0.71

 Nero

-0.68

ITED

-0.65

RAY

-0.65

perties

-0.64

 Revelation

-0.63

Sean

-0.62

POSITIVE LOGITS

 mothers

0.86

hip

0.85

chool

0.84

 moms

0.79

folk

0.76

hips

0.75

heses

0.74

paces

0.71

 opting

0.69

opausal

0.69

Activations Density 0.023%

references to mothers and parenting-related topics

No Comments

No Known Activations