INDEX

Explanations

references to locations or entities related to incarceration and underlying societal issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ObjectName

-0.15

Ð¼Ð¾Ð½

-0.15

onis

-0.15

yat

-0.15

.flip

-0.15

#Region

-0.14

å°¾

-0.14

å¯¶

-0.14

èĩº

-0.14

icari

-0.14

POSITIVE LOGITS

Bul

0.16

swick

0.16

bul

0.16

TRANS

0.15

 trans

0.15

cop

0.15

ãĥ³ãĥĩ

0.14

Cop

0.14

eya

0.14

Mit

0.14

Activations Density 0.034%

references to locations or entities related to incarceration and underlying societal issues

No Comments

No Known Activations