INDEX

Explanations

information related to safety and protection

oai_token-act-pair · gpt-3.5-turbo

phrases indicating safety and wellbeing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

*)

-0.91

Â·Â·

-0.72

 theoret

-0.66

 tended

-0.64

velt

-0.62

Newsletter

-0.58

/-

-0.58

)/

-0.56

 depended

-0.56

phr

-0.56

POSITIVE LOGITS

çĶ°

0.64

 TODAY

0.60

ILA

0.60

asca

0.59

 upcoming

0.59

onto

0.57

emaker

0.56

ãĥ¼ãĥĨãĤ£

0.55

Adds

0.54

"#

0.54

Activations Density 1.998%

information related to safety and protection

phrases indicating safety and wellbeing

No Comments

No Known Activations