INDEX

Explanations

proper nouns, particularly names of people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ijd

-0.16

ÐµÐ¾

-0.15

Ã¡Å¾

-0.15

ÐµÐ²Ñĸ

-0.15

obsolete

-0.14

pty

-0.14

å§Ķ

-0.14

oland

-0.13

.misc

-0.13

ripsi

-0.13

POSITIVE LOGITS

tos

0.15

DPR

0.14

 unofficial

0.13

moh

0.13

rien

0.13

OCI

0.13

ories

0.13

ŀ

0.13

OTES

0.13

 Haram

0.13

Activations Density 0.191%

proper nouns, particularly names of people

No Comments

No Known Activations