INDEX

Explanations

references to specific locations and institutions in various contexts

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

 Unloaded

-0.83

ï¸ı

-0.75

htaking

-0.69

iosis

-0.69

cipled

-0.69

ufact

-0.69

 Prosper

-0.68

ortium

-0.66

zbollah

-0.66

amia

-0.65

POSITIVE LOGITS

mates

1.02

mate

1.00

 clerk

1.00

 regulars

0.90

 mates

0.90

 floor

0.86

 brass

0.86

 attendant

0.84

 administrator

0.84

 attendants

0.84

Activations Density 0.329%

references to specific locations and institutions in various contexts

No Comments

No Known Activations