INDEX

Explanations

references to academic and professional expertise

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

AAAAAAAA

-0.83

":[

-0.71

 disadvantage

-0.67

Plex

-0.65

Phys

-0.63

PRESS

-0.62

 whim

-0.61

 isEnabled

-0.61

hhhh

-0.60

 enjoying

-0.59

POSITIVE LOGITS

 litigation

0.81

 campaigns

0.80

 FOIA

0.75

research

0.74

igree

0.73

 studies

0.66

eport

0.65

 research

0.65

 Horowitz

0.64

 Near

0.64

Activations Density 0.140%

references to academic and professional expertise

No Comments

No Known Activations