INDEX

Explanations

specific references to events or occurrences, particularly involving negative or critical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

askell

-0.15

anca

-0.15

hints

-0.15

vy

-0.15

insky

-0.14

ibir

-0.14

agas

-0.14

å¡ĳ

-0.14

 pulse

-0.14

 Pulse

-0.14

POSITIVE LOGITS

erin

0.20

ysi

0.19

phin

0.16

zahl

0.16

pedo

0.15

Ã§ak

0.15

zza

0.15

ãģ£

0.15

anch

0.14

ostel

0.14

Activations Density 0.007%

specific references to events or occurrences, particularly involving negative or critical contexts

No Comments

No Known Activations