INDEX

Explanations

specific terms related to classification and categorization of subjects or entities

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

undai

-0.81

creen

-0.74

sembly

-0.73

acebook

-0.71

rica

-0.70

uckland

-0.69

cale

-0.68

issan

-0.67

azon

-0.66

agascar

-0.65

POSITIVE LOGITS

wise

0.88

 Specific

0.86

 wise

0.79

Unknown

0.77

 Selection

0.77

 Size

0.76

Maker

0.76

 Tags

0.75

 specificity

0.75

 Types

0.73

Activations Density 0.204%

specific terms related to classification and categorization of subjects or entities

No Comments

No Known Activations