INDEX

Explanations

conjunctions and the word "and."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

isu

-0.16

inux

-0.14

opleft

-0.14

lbrace

-0.14

ovsky

-0.14

uvwxyz

-0.14

urg

-0.13

_marshall

-0.13

iaux

-0.13

POSITIVE LOGITS

/or

0.28

rog

0.19

ific

0.18

rogen

0.17

non

0.16

 semi

0.16

 íĺ¹

0.15

jr

0.14

ators

0.14

ogan

0.14

Activations Density 0.224%

conjunctions and the word "and."

No Comments

No Known Activations