INDEX

Explanations

connectors and qualifiers that indicate emphasis or intention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 NavParams

-0.16

istik

-0.16

ewise

-0.16

RelativeTo

-0.15

forget

-0.14

Ð¸ÑĤÐµÐ»ÑĮ

-0.13

aukee

-0.13

 Charge

-0.13

rud

-0.13

áº£y

-0.13

POSITIVE LOGITS

rogen

0.18

#ad

0.15

rog

0.15

 Prosper

0.15

jac

0.14

áº¡ng

0.14

cam

0.14

 indeed

0.14

ecer

0.14

yl

0.14

Activations Density 0.158%

connectors and qualifiers that indicate emphasis or intention

No Comments

No Known Activations