INDEX

Explanations

dependencies in contextual information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

zes

-0.19

ermann

-0.17

abra

-0.15

zek

-0.15

ãĤī

-0.14

 Trend

-0.14

edla

-0.14

echan

-0.13

inka

-0.13

 essence

-0.13

POSITIVE LOGITS

 circumstances

0.28

 whether

0.28

 circumstance

0.26

whether

0.24

age

0.21

 individual

0.20

 luck

0.20

 factors

0.19

 type

0.19

Whether

0.19

Activations Density 0.062%

dependencies in contextual information

No Comments

No Known Activations