INDEX

Explanations

phrases related to initiating, causing, or starting something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating causation or conditions leading to significant outcomes

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

.",

-0.81

?",

-0.70

",

-0.70

orsi

-0.70

!",

-0.64

âĢ¦."

-0.63

(?,

-0.62

 (Â£

-0.61

.?

-0.61

',

-0.60

POSITIVE LOGITS

ãĥĩãĤ£

0.78

ãĥ¥

0.74

 winds

0.66

ãĥĭ

0.62

voy

0.62

 VIDE

0.60

arently

0.59

ãĥĨãĤ£

0.58

arks

0.58

ãĥİ

0.56

Activations Density 0.464%

phrases related to initiating, causing, or starting something

phrases indicating causation or conditions leading to significant outcomes

No Comments

No Known Activations