INDEX

Explanations

terms related to summation or adding together values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 criticised

-0.14

nam

-0.13

tob

-0.13

 youre

-0.13

utzer

-0.13

ØŃÙĩ

-0.13

AREN

-0.13

_areas

-0.13

POSITIVE LOGITS

awy

0.16

iola

0.15

 Narrative

0.15

NYT

0.15

NY

0.15

 theater

0.14

exus

0.14

 Noon

0.13

 supportive

0.13

xygen

0.13

Activations Density 0.000%

terms related to summation or adding together values

No Comments

No Known Activations