INDEX

Explanations

instances of the word "change" and its context in various discussions

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

pots

-0.73

tips

-0.72

soever

-0.71

warn

-0.71

rs

-0.70

YES

-0.69

lehem

-0.69

linger

-0.66

isters

-0.66

WOOD

-0.65

POSITIVE LOGITS

 favor

1.01

 fortunes

1.01

 attitude

0.99

 tone

0.97

 attitudes

0.95

 consciousness

0.95

 perception

0.94

 demeanor

0.87

 favour

0.87

 mindset

0.87

Activations Density 0.040%

instances of the word "change" and its context in various discussions

No Comments

No Known Activations