INDEX

Explanations

notations or labels indicating important notes or warnings

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

 manif

-0.71

 eroded

-0.68

izons

-0.65

orate

-0.65

onet

-0.65

ior

-0.64

 trou

-0.64

etheless

-0.63

aled

-0.63

 outwe

-0.63

POSITIVE LOGITS

Due

0.99

 Contains

0.94

 Whilst

0.92

 This

0.91

 Originally

0.91

 Restaur

0.90

 Beware

0.88

 Certain

0.87

 Unless

0.86

 Sorry

0.85

Activations Density 0.042%

notations or labels indicating important notes or warnings

No Comments

No Known Activations