INDEX

Explanations

specific patterns or elements related to numerical identifiers and encoding styles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

kyt

-0.16

-0.15

ÑīÐ¸Ð½

-0.15

Ŀ

-0.14

POV

-0.14

ä¼Ĺ

-0.14

acier

-0.14

/legal

-0.14

jud

-0.14

 Turn

-0.14

POSITIVE LOGITS

OLA

0.17

 Wheeler

0.17

 cá»¥c

0.17

ola

0.16

Nick

0.15

 Nick

0.15

curso

0.15

icy

0.14

ugu

0.14

 lyon

0.14

Activations Density 0.029%

specific patterns or elements related to numerical identifiers and encoding styles

No Comments

No Known Activations