INDEX

Explanations

words related to measurements or quantities

oai_token-act-pair · gpt-3.5-turbo

terms related to specific food or drink products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ï¿½

-0.64

 ..........

-0.63

-0.62

 âĢİ

-0.59

 ï¿½

-0.56

ï¿½ï¿½

-0.55

``

-0.54

âĸł

-0.51

 âĢº

-0.50

-0.49

POSITIVE LOGITS

 Annotations

0.56

pload

0.54

 guiActiveUn

0.53

erker

0.51

uci

0.51

pedia

0.51

ultimate

0.51

gyn

0.50

ghazi

0.50

 Hera

0.50

Activations Density 3.515%

words related to measurements or quantities

terms related to specific food or drink products

No Comments

No Known Activations