INDEX

Explanations

instances of emphasis or focal points within a narrative

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

gary

-0.15

okie

-0.14

[section

-0.14

izzo

-0.14

CompleteListener

-0.14

vid

-0.14

Kaw

-0.13

 Pied

-0.13

fx

-0.13

ousse

-0.13

POSITIVE LOGITS

ypad

0.15

lies

0.14

regon

0.14

 representation

0.14

 Sketch

0.14

ÚĺÙĩ

0.14

Ã¶

0.13

çĦ¼

0.13

æĹ¥

0.13

_MAGIC

0.13

Activations Density 0.001%

instances of emphasis or focal points within a narrative

No Comments

No Known Activations