INDEX

Explanations

actions related to events, performances, and activities

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

oud

-0.14

/or

-0.14

kel

-0.14

 Matters

-0.13

 matters

-0.13

orex

-0.13

angkan

-0.13

ÐµÑĩ

-0.13

figcaption

-0.13

ault

-0.13

POSITIVE LOGITS

ednou

0.16

-thumbnails

0.15

ehler

0.15

ubes

0.14

Charsets

0.14

erde

0.14

iele

0.14

ãĢ

0.13

_SOL

0.13

oulder

0.13

Activations Density 0.325%

actions related to events, performances, and activities

No Comments

No Known Activations