INDEX

Explanations

characters and their interactions in a narrative context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 confir

-0.88

 ballpark

-0.84

 credential

-0.79

 Univ

-0.77

 Qualcomm

-0.76

 enrichment

-0.75

 weekday

-0.74

 productivity

-0.74

 technical

-0.74

 industrial

-0.74

POSITIVE LOGITS

ilda

1.26

abeth

1.16

wen

1.15

ilyn

1.13

asha

1.07

ako

1.06

uko

1.06

iona

1.05

ora

1.05

elia

1.03

Activations Density 0.266%

characters and their interactions in a narrative context

No Comments

No Known Activations