INDEX

Explanations

unique identifiers or terms related to specific titles or characters from media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

bedo

-0.17

aminer

-0.17

Bes

-0.15

ãĤŃãĥ³ãĤ°

-0.14

 Bust

-0.14

oples

-0.14

Ã´

-0.14

Č↵

-0.14

Tib

-0.13

 Holding

-0.13

POSITIVE LOGITS

 Spiel

0.17

 Ãºn

0.15

íķĺ

0.15

mÃ¡

0.14

ìĿ´

0.14

jong

0.14

Ð´Ð°Ð¿

0.14

æ¿Ł

0.14

¤í

0.14

 Korea

0.14

Activations Density 0.001%

unique identifiers or terms related to specific titles or characters from media

No Comments

No Known Activations