INDEX

Explanations

references to physical movement or ascension

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

Closure

-0.64

Synopsis

-0.63

ums

-0.61

uel

-0.60

ATA

-0.59

 understatement

-0.58

Reviewed

-0.56

pers

-0.56

redo

-0.56

 suff

-0.56

POSITIVE LOGITS

stairs

1.05

rights

1.04

 stairs

1.00

stage

0.96

river

0.92

raised

0.91

grading

0.86

 hills

0.82

dating

0.80

grades

0.80

Activations Density 0.029%

references to physical movement or ascension

No Comments

No Known Activations