INDEX

Explanations

references to specific harbors

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

DonaldTrump

-0.82

uable

-0.79

hesis

-0.78

iferation

-0.72

Downloadha

-0.71

ieve

-0.71

ctive

-0.69

yson

-0.68

iosis

-0.67

otine

-0.66

POSITIVE LOGITS

front

0.90

 Cruise

0.84

 Ship

0.82

 Harbour

0.79

 Shore

0.79

 Yard

0.78

ships

0.75

 Ships

0.73

ward

0.73

 Harbor

0.72

Activations Density 0.007%

references to specific harbors

No Comments

No Known Activations