INDEX

Explanations

proper nouns followed by possessive markers

oai_token-act-pair · gpt-3.5-turbo

the presence of end-of-text markers or similar empty structures indicating separation in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

 visitation

-0.73

 precon

-0.64

 organizing

-0.63

ãĥ¼ãĥĨ

-0.63

 organised

-0.62

 UNESCO

-0.61

CPR

-0.61

 subcontract

-0.60

 kinderg

-0.58

 organized

-0.58

POSITIVE LOGITS

lightly

1.12

orted

1.07

ources

1.06

atisf

1.06

aved

1.06

omew

1.05

olo

1.05

ESSION

1.00

atellite

0.98

iblings

0.97

Activations Density 0.045%

proper nouns followed by possessive markers

the presence of end-of-text markers or similar empty structures indicating separation in the text

No Comments

No Known Activations