INDEX

Explanations

reports of violent or brutal acts

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

 Bowman

-0.18

razier

-0.16

trace

-0.15

éĺħè¯»

-0.14

Trace

-0.14

çĿ£

-0.14

defs

-0.14

æİĽ

-0.14

omentum

-0.14

Scan

-0.13

POSITIVE LOGITS

 reporting

0.45

 report

0.43

 reports

0.42

 Reporting

0.39

reports

0.38

-report

0.36

Reporting

0.36

 Reports

0.36

report

0.35

 REPORT

0.34

Activations Density 0.236%

reports of violent or brutal acts

No Comments

No Known Activations