INDEX

Explanations

phrases that indicate a transition or change in state or time

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

 zwar

-0.18

ossier

-0.17

.appspot

-0.16

heimer

-0.15

olini

-0.14

imary

-0.14

.jboss

-0.14

ãģ«ãĤĪ

-0.14

amaz

-0.14

/pg

-0.13

POSITIVE LOGITS

 ultimately

0.25

 being

0.25

 Ultimately

0.20

 eventually

0.20

 finally

0.19

being

0.18

 Being

0.17

Being

0.17

 becoming

0.17

 then

0.17

Activations Density 0.044%

phrases that indicate a transition or change in state or time

No Comments

No Known Activations