INDEX

Explanations

references to symbols or representations of significant ideas or people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

onders

-0.18

IGHL

-0.16

wick

-0.15

ylie

-0.15

ylon

-0.15

MMdd

-0.15

enia

-0.14

é±

-0.14

eenth

-0.14

Ã¼st

-0.14

POSITIVE LOGITS

nect

0.24

oc

0.20

hole

0.17

odule

0.17

nection

0.16

esson

0.15

izes

0.15

duct

0.15

owie

0.15

ography

0.15

Activations Density 0.027%

references to symbols or representations of significant ideas or people

No Comments

No Known Activations