INDEX

Explanations

phrases that describe relationships or connections among subjects

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

lement

-0.81

SI

-0.80

cies

-0.78

ento

-0.77

igun

-0.76

é¾įåĸļå£«

-0.75

orie

-0.74

NET

-0.74

va

-0.74

INA

-0.74

POSITIVE LOGITS

 ordinary

0.80

 adulthood

0.76

 typical

0.75

 predecessors

0.74

 modern

0.74

 habitual

0.69

 sophistication

0.69

 infancy

0.68

Sod

0.67

 adolescence

0.67

Activations Density 0.076%

phrases that describe relationships or connections among subjects

No Comments

No Known Activations