INDEX

Explanations

references to advocacy or attempts to influence change

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

 Scotia

-0.71

istically

-0.67

 Gleaming

-0.66

 concess

-0.66

 richness

-0.66

racuse

-0.64

 splend

-0.63

BuyableInstoreAndOnline

-0.63

ocally

-0.63

 Exile

-0.62

POSITIVE LOGITS

back

1.29

cart

1.10

ups

1.05

backs

1.03

chairs

0.99

button

0.95

overs

0.95

down

0.92

BACK

0.89

bike

0.89

Activations Density 0.012%

references to advocacy or attempts to influence change

No Comments

No Known Activations