INDEX

Explanations

key terms related to technology and textual elements like section headers

oai_token-act-pair · gpt-3.5-turbo

key terms associated with events and significant topics in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 respectively

-0.65

)."

-0.64

 thereof

-0.62

]."

-0.59

 therein

-0.58

omever

-0.58

terday

-0.57

Â·Â·

-0.54

åĳ

-0.53

}.

-0.53

POSITIVE LOGITS

 Releases

0.70

 Profile

0.66

 âĵĺ

0.65

emo

0.62

 Edit

0.60

 Sets

0.60

 Scale

0.58

 Characters

0.58

 Manager

0.57

 Gets

0.57

Activations Density 1.005%

key terms related to technology and textual elements like section headers

key terms associated with events and significant topics in various contexts

No Comments

No Known Activations