INDEX

Explanations

variations of the word "resolve" and its forms 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

ingham

-0.81

 helicop

-0.66

antha

-0.66

tu

-0.66

accessible

-0.61

phies

-0.61

 waived

-0.60

 performed

-0.59

 Signed

-0.59

 tabl

-0.58

POSITIVE LOGITS

 revolves

1.00

 revolving

0.90

 around

0.87

 solely

0.86

 centered

0.83

 loosely

0.82

actionDate

0.81

 principally

0.77

 primarily

0.76

 mainly

0.76

Activations Density 0.020%

variations of the word "resolve" and its forms in different contexts

No Comments

No Known Activations