INDEX

Explanations

invocations to subscribe to newsletters or mailing lists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

eger

-0.17

atte

-0.16

ise

-0.16

 obliv

-0.15

obe

-0.15

usc

-0.15

enz

-0.14

ald

-0.14

rys

-0.14

istry

-0.14

POSITIVE LOGITS

/un

0.16

Ð¶Ð½Ð¾

0.15

ificantly

0.15

ãĥ¼ãĥĦ

0.15

_globals

0.15

åĪ°çļĦ

0.15

 ngoáº¡i

0.15

olor

0.15

noop

0.14

=sub

0.14

Activations Density 0.011%

invocations to subscribe to newsletters or mailing lists

No Comments

No Known Activations