INDEX

Explanations

references to specific locations and relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

patch

-0.19

 patch

-0.18

 Patch

-0.15

rouw

-0.15

Ord

-0.15

Arr

-0.15

Patch

-0.14

Hoy

-0.14

ÐµÑĢÐµÐ´

-0.14

afen

-0.14

POSITIVE LOGITS

 nail

0.18

 nose

0.18

ima

0.18

 Ð´Ð¾Ð»

0.18

 rodi

0.17

se

0.17

 Ð¼Ð¾Ð¶Ðµ

0.17

 Ð´ÐµÐ»Ðµ

0.17

mo

0.17

 Ð½Ð°Ð»

0.16

Activations Density 0.002%

references to specific locations and relationships

No Comments

No Known Activations