INDEX

Explanations

concepts related to social conflict and violence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 amplify

-0.15

ISTR

-0.14

.descriptor

-0.14

 absorbing

-0.13

iska

-0.13

flt

-0.13

inue

-0.13

pill

-0.13

andon

-0.13

 attempted

-0.13

POSITIVE LOGITS

trigger

0.36

prompt

0.35

 trigger

0.35

 prompt

0.35

 Trigger

0.33

 prompts

0.33

 catal

0.32

Prompt

0.32

 motivate

0.32

 Prompt

0.32

Activations Density 0.725%

concepts related to social conflict and violence

No Comments

No Known Activations