INDEX

Explanations

phrases indicating attempts to communicate or connect with others

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rey

-0.17

erland

-0.17

aily

-0.17

aldo

-0.16

 Scrap

-0.14

nok

-0.14

Tes

-0.14

rollers

-0.14

aday

-0.14

 statement

-0.14

POSITIVE LOGITS

recated

0.17

ãĥ¼ãĥĨ

0.17

zcze

0.16

uset

0.16

olson

0.15

.idea

0.15

IPC

0.15

atz

0.15

wdx

0.14

umhur

0.14

Activations Density 0.348%

phrases indicating attempts to communicate or connect with others

No Comments

No Known Activations