INDEX

Explanations

phrases indicating relationships between people or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

-0.15

ëŀĢ

-0.15

redo

-0.15

ICC

-0.14

ety

-0.14

att

-0.13

rei

-0.13

uh

-0.13

 INTERRUPTION

-0.13

POSITIVE LOGITS

arding

0.19

Ð¾Ð½Ð¾Ð²

0.18

aeper

0.17

ilden

0.16

ungan

0.15

 vids

0.15

antium

0.15

 ours

0.14

ãĢĪ

0.14

 hala

0.14

Activations Density 0.339%

phrases indicating relationships between people or entities

No Comments

No Known Activations