INDEX

Explanations

references to financial wrongdoing and social welfare issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ERO

-0.16

oric

-0.15

ILI

-0.14

eya

-0.14

enga

-0.14

/inet

-0.14

habi

-0.14

 Mons

-0.14

ycz

-0.14

sessionId

-0.13

POSITIVE LOGITS

TOTYPE

0.17

 implementation

0.17

osti

0.16

IMPLEMENT

0.15

 Kear

0.15

 Trophy

0.15

 Dana

0.15

 æ¶

0.15

SK

0.15

 Implementation

0.15

Activations Density 0.001%

references to financial wrongdoing and social welfare issues

No Comments

No Known Activations