INDEX

Explanations

punctuation and formatting elements in the text

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

 downs

-0.17

 revis

-0.15

 personalised

-0.15

av

-0.15

 discrim

-0.14

nud

-0.14

inst

-0.14

zik

-0.14

 Ð¿ÑĢÐ¾ÑĦÐµÑģÑģÐ¸Ð¾Ð½Ð°Ð»ÑĮ

-0.13

POSITIVE LOGITS

 become

0.15

eph

0.15

arr

0.14

Use

0.14

eger

0.14

Use

0.14

 Neuroscience

0.14

Use

0.14

 conduct

0.13

ãĥªãĥ¼

0.13

Activations Density 0.117%

punctuation and formatting elements in the text

No Comments

No Known Activations