INDEX

Explanations

concepts related to representation in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ery

-0.18

reich

-0.18

otropic

-0.18

ÌĢ

-0.17

ral

-0.16

Ø³ØªØ§ÙĨ

-0.15

åĢ

-0.15

Ð»ÑıÐµÑĤ

-0.15

erty

-0.15

jar

-0.15

POSITIVE LOGITS

 ÑģÐ¾Ð±Ð¾Ð¹

0.20

atively

0.18

Ãºblica

0.15

æĥł

0.15

ÑĥÐ¶

0.14

bens

0.14

Represent

0.14

ational

0.14

Fizz

0.13

atives

0.13

Activations Density 0.018%

concepts related to representation in various contexts

No Comments

No Known Activations