INDEX

Explanations

various forms of expressions or idioms indicating comparison or similarity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ourcem

-0.16

iferay

-0.15

ä¸¸

-0.14

 Ekon

-0.14

 tear

-0.14

Ã´te

-0.13

XE

-0.13

uese

-0.13

bon

-0.13

POSITIVE LOGITS

leigh

0.16

auga

0.16

acey

0.15

ugen

0.15

lt

0.15

-proof

0.15

uzzi

0.14

 accr

0.13

umpt

0.13

Activations Density 0.143%

various forms of expressions or idioms indicating comparison or similarity

No Comments

No Known Activations