INDEX

Explanations

expressions of apology and regret

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

íķĳ

-0.15

odge

-0.14

Î¸Î±

-0.14

 Observatory

-0.14

MMdd

-0.14

 Minds

-0.13

ãĥ³ãĥģ

-0.13

adder

-0.13

ador

-0.13

etwork

-0.13

POSITIVE LOGITS

SENS

0.16

_ctx

0.15

 Ideal

0.15

apus

0.14

alin

0.14

ÑĥÐºÐ¾Ð²

0.14

 meant

0.14

 ideal

0.14

 privileged

0.14

 Priv

0.14

Activations Density 0.053%

expressions of apology and regret

No Comments

No Known Activations