INDEX

Explanations

proper names or surnames

oai_token-act-pair · gpt-3.5-turbo

frequent names or words related to personal identification

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

ãĤ´ãĥ³

-0.84

ħĭ

-0.74

ĸļ

-0.72

 backlog

-0.66

Õ

-0.66

Ļ½

-0.65

DAQ

-0.65

wcs

-0.64

İĭ

-0.63

çīĪ

-0.62

POSITIVE LOGITS

ruary

1.13

ascript

0.95

renheit

0.93

gerald

0.85

owship

0.83

illet

0.79

ilon

0.76

ilies

0.76

acci

0.75

ilial

0.73

Activations Density 0.089%

proper names or surnames

frequent names or words related to personal identification

No Comments

No Known Activations