INDEX

Explanations

procedural steps and instructions

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

jong

-0.15

EventManager

-0.14

zier

-0.14

ä¸Ī

-0.14

addtogroup

-0.14

ANTE

-0.14

raud

-0.13

ä¸Ģäºº

-0.13

uli

-0.13

ÑģÑĤÑĭ

-0.13

POSITIVE LOGITS

 step

0.20

 then

0.16

 Step

0.15

 STEP

0.15

 Then

0.15

á»³

0.15

Step

0.15

0.14

step

0.14

 steps

0.14

Activations Density 0.049%

procedural steps and instructions

No Comments

No Known Activations