INDEX

Explanations

scientific and technical terms related to research, equipment, medical procedures, and procedures

oai_token-act-pair · gpt-3.5-turbo

terms related to scientific processes and research methodologies

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

partisan

-0.79

intend

-0.61

isode

-0.61

govtrack

-0.59

netflix

-0.59

 Founders

-0.58

 Fiscal

-0.58

atform

-0.58

arcity

-0.57

 DeVos

-0.57

POSITIVE LOGITS

 guiActiveUn

0.84

ãĤ©

0.75

natureconservancy

0.58

herty

0.57

arily

0.56

emonium

0.56

çĶ°

0.56

istine

0.55

}}}

0.54

 symbol

0.54

Activations Density 1.687%

scientific and technical terms related to research, equipment, medical procedures, and procedures

terms related to scientific processes and research methodologies

No Comments

No Known Activations