INDEX

Explanations

keywords related to specific political or cultural topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

.PreparedStatement

-0.17

 instead

-0.17

instead

-0.16

*M

-0.14

 à¤¨

-0.14

ÛĮÙĨÙĩ

-0.14

Çİ

-0.14

olas

-0.13

.reducer

-0.13

draul

-0.13

POSITIVE LOGITS

ÂłT

0.17

 Ð¢

0.16

 Schwarz

0.15

 ãĤ¦

0.15

alse

0.15

TAS

0.15

 Î¤

0.15

se

0.14

 sequ

0.14

Activations Density 0.028%

keywords related to specific political or cultural topics

No Comments

No Known Activations