INDEX

Explanations

non-specific references to people or general statements about individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

uki

-0.17

uj

-0.15

 blackmail

-0.15

ichtig

-0.15

mou

-0.15

zik

-0.15

pog

-0.15

isan

-0.15

rous

-0.14

esty

-0.14

POSITIVE LOGITS

leh

0.18

rex

0.14

XCT

0.14

ervoir

0.13

@nate

0.13

HQ

0.13

.AF

0.13

GetInt

0.13

oric

0.13

Activations Density 0.000%

non-specific references to people or general statements about individuals

No Comments

No Known Activations