INDEX

Explanations

phrases related to recommendations and suggestions

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

ipel

-0.75

ophon

-0.73

zona

-0.73

seconds

-0.72

arton

-0.71

riors

-0.70

ccording

-0.70

yrs

-0.68

hazard

-0.66

ansk

-0.65

POSITIVE LOGITS

 suggestions

0.80

 suggestion

0.75

 sugg

0.75

 guideline

0.74

 recommendations

0.71

 recommendation

0.70

osal

0.70

ariat

0.70

 criteria

0.69

 regarding

0.69

Activations Density 0.015%

phrases related to recommendations and suggestions

No Comments

No Known Activations