INDEX

Explanations

time indicators

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

emouth

-0.16

edium

-0.16

oten

-0.15

restriction

-0.15

bz

-0.15

opensource

-0.15

åħ¥ãĤĬ

-0.14

umont

-0.14

AsyncResult

-0.14

eda

-0.14

POSITIVE LOGITS

ERICAN

0.20

EDT

0.19

azing

0.16

PDT

0.16

inton

0.16

PST

0.15

hare

0.15

 Caller

0.14

ET

0.14

âķĳ

0.14

Activations Density 0.009%

time indicators

No Comments

No Known Activations