INDEX

Explanations

discussions about Jewish identity and denominations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hea

-0.17

prot

-0.17

'gc

-0.15

uslim

-0.14

anse

-0.14

efon

-0.14

agt

-0.14

anst

-0.14

ç§

-0.13

 prote

-0.13

POSITIVE LOGITS

 Orth

0.33

 observ

0.32

Orth

0.29

 Observ

0.28

observ

0.28

 orth

0.26

hal

0.23

mit

0.23

 Conservative

0.22

Observ

0.22

Activations Density 0.047%

discussions about Jewish identity and denominations

No Comments

No Known Activations