INDEX

Explanations

repeated punctuation marks, especially commas

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ONSORED

-0.78

rea

-0.75

izon

-0.74

=/

-0.72

ITNESS

-0.72

=(

-0.71

ESE

-0.70

Tools

-0.70

 Journals

-0.69

=~

-0.69

POSITIVE LOGITS

 rectangular

0.94

 albeit

0.90

um

0.87

 juicy

0.86

 rainy

0.85

 colourful

0.84

 sandy

0.84

 curly

0.83

 spherical

0.82

 shallow

0.81

Activations Density 0.058%

repeated punctuation marks, especially commas

No Comments

No Known Activations