INDEX

Explanations

phrases that introduce individuals, particularly those involved in specific roles or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

Ã¤Ã¤n

-0.14

 chaired

-0.14

ÏħÏĢ

-0.13

Ã¤nner

-0.13

_CPP

-0.13

antha

-0.13

ãĥ³ãĤ¯

-0.12

eck

-0.12

 Armed

-0.12

uploaded

-0.12

POSITIVE LOGITS

recent

0.21

 recently

0.21

 served

0.18

 recent

0.17

0.16

 helped

0.16

à¹Ģà¸Ľ

0.15

isa

0.15

 serve

0.15

 serves

0.15

Activations Density 0.151%

phrases that introduce individuals, particularly those involved in specific roles or events

No Comments

No Known Activations