INDEX

Explanations

references to hierarchical structures or titles related to authority or accomplishment

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

ABE

-0.73

 Kamp

-0.72

ORTS

-0.67

TPS

-0.66

uana

-0.65

 Citation

-0.65

 Dragonbound

-0.63

 Bach

-0.61

ensation

-0.61

Soy

-0.60

POSITIVE LOGITS

ipel

1.38

itect

1.15

bishop

1.13

ival

1.10

ivist

0.99

adia

0.98

rival

0.96

aic

0.95

di

0.93

iving

0.84

Activations Density 0.006%

references to hierarchical structures or titles related to authority or accomplishment

No Comments

No Known Activations