INDEX

Explanations

references to music, film, and popular culture names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ENSE

-0.13

ffffff

-0.13

uyla

-0.13

NIL

-0.13

 Levy

-0.13

plementation

-0.13

 Gros

-0.13

Haj

-0.13

å¹³

-0.12

 Hugo

-0.12

POSITIVE LOGITS

Ð¶Ð°

0.14

-ÐĴ

0.14

ampoline

0.13

-Am

0.13

Ð¾Ð²Ð°Ð½Ð¸

0.13

\Blueprint

0.12

ppelin

0.12

ă

0.12

Äļ

0.12

-Cal

0.12

Activations Density 0.488%

references to music, film, and popular culture names and titles

No Comments

No Known Activations

references to music, film, and popular culture names and titles

No Comments

No Known Activations