INDEX

Explanations

language or terms related to communication and accessibility for a general audience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

DDL

-0.16

amb

-0.15

Ãª

-0.14

 Rosenstein

-0.14

olkien

-0.14

reed

-0.14

eton

-0.14

ÄĽ

-0.14

cko

-0.14

 substitution

-0.13

POSITIVE LOGITS

MOVED

0.17

AREST

0.16

IMATE

0.15

UBLE

0.15

ARIANT

0.15

\@

0.14

ĤŃ

0.14

 science

0.14

ROUGH

0.14

 excess

0.14

Activations Density 0.253%

language or terms related to communication and accessibility for a general audience

No Comments

No Known Activations