INDEX

Explanations

references to various forms of labor or tasks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

oodles

-0.17

Č↵

-0.15

riel

-0.15

urger

-0.15

LOPT

-0.15

oodle

-0.15

erosis

-0.14

enton

-0.14

uzu

-0.14

POSITIVE LOGITS

gan

0.15

 Vern

0.15

ysl

0.14

/Documents

0.14

plet

0.14

Scheme

0.14

 mechanic

0.14

.assets

0.13

earn

0.13

eren

0.13

Activations Density 0.097%

references to various forms of labor or tasks

No Comments

No Known Activations