INDEX

Explanations

instances of significant actions or events related to failures or decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

EMPLARY

-0.15

addtogroup

-0.15

ersed

-0.14

ç«ĭãģ¡

-0.14

steder

-0.14

aktu

-0.13

:;"

-0.13

 beden

-0.13

idor

-0.13

ieÅ¼

-0.13

POSITIVE LOGITS

 Space

0.19

 spacecraft

0.18

 nomin

0.16

.space

0.16

 NASA

0.16

 space

0.15

↵

0.15

space

0.14

 Ches

0.14

Space

0.14

Activations Density 0.000%

instances of significant actions or events related to failures or decisions

No Comments

No Known Activations