INDEX

Explanations

structured question-answer formats and indicators of a discussion or inquiry

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

guard

-0.13

niÅ¾

-0.13

ses

-0.13

/as

-0.13

.fm

-0.12

stal

-0.12

 terminator

-0.12

anske

-0.12

 latter

-0.12

Dy

-0.12

POSITIVE LOGITS

0.16

avy

0.14

 there

0.14

AREST

0.14

ahlen

0.14

There

0.13

 There

0.13

ximo

0.13

_KHR

0.13

ivÄĽ

0.13

Activations Density 0.048%

structured question-answer formats and indicators of a discussion or inquiry

No Comments

No Known Activations