INDEX

Explanations

expressions related to blame and conspiracy theories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ulti

-0.14

HttpResponse

-0.14

bias

-0.14

çĮ®

-0.14

 conduct

-0.13

inka

-0.13

åľŃ

-0.13

rade

-0.13

ãĥ«ãĥĪ

-0.13

POSITIVE LOGITS

 casting

0.15

 perceived

0.15

 convinced

0.15

 Paste

0.15

-Ta

0.14

 conspiracy

0.14

 targets

0.14

èĸ

0.14

unp

0.13

mag

0.13

Activations Density 0.131%

expressions related to blame and conspiracy theories

No Comments

No Known Activations