INDEX

Explanations

references to possessive pronouns indicating ownership, particularly focusing on "their" and "his"

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

 Berk

-0.71

 Frie

-0.70

Byr

-0.70

 Seym

-0.69

 Barrett

-0.68

å§

-0.65

 Niet

-0.64

 Edison

-0.64

Kuh

-0.62

 Strauss

-0.61

POSITIVE LOGITS

own

0.79

 favourite

0.75

 favorite

0.74

voice

0.70

bably

0.69

metics

0.68

 finest

0.67

 sqor

0.66

elf

0.66

osponsors

0.66

Activations Density 0.036%

references to possessive pronouns indicating ownership, particularly focusing on "their" and "his"

No Comments

No Known Activations