INDEX

Explanations

references to programs and initiatives, particularly in relation to timing and actions

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

\\\\\\\\

-0.72

ãĤ´ãĥ³

-0.71

gall

-0.70

mson

-0.66

tein

-0.65

ãĥķãĤ©

-0.61

tain

-0.60

apego

-0.60

istani

-0.59

forts

-0.58

POSITIVE LOGITS

 Duration

0.81

 thereafter

0.77

Duration

0.73

 duration

0.69

arate

0.65

duration

0.63

aturdays

0.62

ummies

0.61

 Seller

0.61

ipolar

0.61

Activations Density 0.003%

references to programs and initiatives, particularly in relation to timing and actions

No Comments

No Known Activations