INDEX

Explanations

criticisms of media and entertainment

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

 sÃ¡t

-0.15

assis

-0.15

esser

-0.15

Ã¦k

-0.15

dea

-0.14

íĹ

-0.14

esa

-0.14

 Hann

-0.14

ativos

-0.14

utorial

-0.14

POSITIVE LOGITS

 decent

0.17

 SOME

0.16

 attempt

0.16

 nicely

0.16

 undeniable

0.15

 useful

0.15

 Attempt

0.15

 nice

0.15

nic

0.15

nice

0.15

Activations Density 0.322%

criticisms of media and entertainment

No Comments

No Known Activations