INDEX

Explanations

phrases indicating personal understanding and analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

idal

-0.16

ieux

-0.16

ream

-0.15

ieu

-0.14

eba

-0.14

ffer

-0.14

ãģĹãģ¾

-0.14

typeName

-0.14

 doub

-0.14

hil

-0.14

POSITIVE LOGITS

 gather

0.27

 gathered

0.25

 Gather

0.23

 gathers

0.23

 gathering

0.22

gather

0.20

 understand

0.20

 Gathering

0.20

 understanding

0.18

 understands

0.18

Activations Density 0.035%

phrases indicating personal understanding and analysis

No Comments

No Known Activations