INDEX

Explanations

phrases related to conditions and expected outcomes in tests

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ÏĢÎ¬

-0.15

Ø³ÛĮ

-0.14

achs

-0.14

indexed

-0.14

 Ø§Ùĩ

-0.14

ÑĭÑĪ

-0.13

rets

-0.13

ardon

-0.13

hee

-0.13

Ä±lÄ±

-0.13

POSITIVE LOGITS

ffen

0.16

eken

0.15

.sky

0.15

legen

0.14

ackbar

0.14

ISON

0.14

ISCO

0.14

richt

0.13

ison

0.13

vision

0.13

Activations Density 0.006%

phrases related to conditions and expected outcomes in tests

No Comments

No Known Activations