INDEX

Explanations

patterns in numerical codes or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

åŁŁ

-0.16

-0.15

ney

-0.15

ahu

-0.14

ani

-0.14

las

-0.14

Ã¦k

-0.14

hel

-0.13

 pepper

-0.13

POSITIVE LOGITS

ertino

0.18

_TERM

0.17

.jd

0.16

áº½

0.15

ëĭµ

0.15

.Guna

0.15

 ÑĢÐ¾Ð±

0.15

rrha

0.15

orra

0.14

ç»

0.14

Activations Density 0.010%

patterns in numerical codes or identifiers

No Comments

No Known Activations