INDEX

Explanations

phrases that indicate the initiation or beginning of actions or processes

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

-0.15

anders

-0.15

 thinking

-0.15

ander

-0.14

stone

-0.14

 overcome

-0.14

stones

-0.14

vents

-0.14

andering

-0.13

audi

-0.13

POSITIVE LOGITS

 happening

0.19

zia

0.19

oad

0.17

áº¥m

0.17

 being

0.16

 humble

0.16

oshi

0.15

 appearing

0.15

appen

0.14

Its

0.14

Activations Density 0.052%

phrases that indicate the initiation or beginning of actions or processes

No Comments

No Known Activations