INDEX

Explanations

instances of the word "trump" and its variations in different 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

Interstitial

-0.74

 Norn

-0.71

 Argon

-0.69

 Relief

-0.66

 Ecology

-0.65

 Publisher

-0.64

ASED

-0.63

 Rite

-0.63

 alienation

-0.63

 Lauder

-0.62

POSITIVE LOGITS

eting

1.58

eters

1.21

ets

1.11

eter

1.09

 trump

0.99

stakes

0.99

ed

0.98

et

0.94

ett

0.94

essed

0.88

Activations Density 0.005%

instances of the word "trump" and its variations in different contexts

No Comments

No Known Activations