INDEX

Explanations

connections between different concepts or actions

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

si

-0.17

icket

-0.15

etooth

-0.15

Î¸ÎµÎ½

-0.15

/Dk

-0.14

éĻ

-0.14

cela

-0.14

ByID

-0.14

StateManager

-0.14

mine

-0.13

POSITIVE LOGITS

 hence

0.20

 same

0.16

 also

0.16

Moreover

0.15

 Hence

0.15

 Hawth

0.14

 Same

0.14

 slightest

0.14

ZE

0.14

rv

0.14

Activations Density 0.228%

connections between different concepts or actions

No Comments

No Known Activations