INDEX

Explanations

specific names or terms within a text

oai_token-act-pair · gpt-3.5-turbo

specific names, events, or references related to notable cultural or technological topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ï¿½

-0.63

 ï¿½

-0.58

 ..........

-0.57

ï¿½ï¿½

-0.56

 âĢİ

-0.55

``

-0.51

``

-0.49

.--

-0.48

âĸł

-0.47

POSITIVE LOGITS

 Annotations

0.61

uci

0.54

pload

0.51

shapeshifter

0.51

schild

0.50

 eclipse

0.49

 guiActiveUn

0.49

ghazi

0.47

ictive

0.47

Else

0.47

Activations Density 4.058%

specific names or terms within a text

specific names, events, or references related to notable cultural or technological topics

No Comments

No Known Activations