INDEX

Explanations

instances of interaction or engagement 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 @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

/from

-0.17

OfMonth

-0.17

anou

-0.15

ovan

-0.14

/of

-0.14

ication

-0.13

bre

-0.13

urus

-0.13

readcr

-0.13

vention

-0.13

POSITIVE LOGITS

 with

0.16

 upon

0.16

on

0.15

ä¸Ģä¸ĭ

0.15

OffsetTable

0.15

GuidId

0.15

ëĸ

0.15

ingly

0.14

/testify

0.14

iu

0.14

Activations Density 0.580%

instances of interaction or engagement in various contexts

No Comments

No Known Activations