INDEX

Explanations

numeric values associated with years

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

Margins

-0.15

jni

-0.15

imary

-0.14

hti

-0.14

iere

-0.14

quota

-0.14

ury

-0.13

YLE

-0.13

disabled

-0.13

agine

-0.13

POSITIVE LOGITS

 Shib

0.16

 Rico

0.15

/gtest

0.14

lander

0.14

çħĻ

0.14

GLOSS

0.13

rico

0.13

ä¿

0.13

.NewLine

0.13

balls

0.13

Activations Density 0.018%

numeric values associated with years

No Comments

No Known Activations