INDEX

Explanations

references to dishonesty and falsehoods in political discourse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ÑĪÐºÐ¸

-0.17

iba

-0.15

ÑĤÐ¾ÑĤ

-0.14

ÐµÑģÐ¸

-0.14

 åº

-0.14

 Stub

-0.14

ÑĢÐ¾Ñī

-0.14

_hooks

-0.13

uctor

-0.13

POSITIVE LOGITS

 Perkins

0.14

CJK

0.14

ae

0.14

dex

0.14

_gettime

0.13

 Oaks

0.13

atto

0.13

 porr

0.13

 inconvenience

0.13

Activations Density 0.148%

references to dishonesty and falsehoods in political discourse

No Comments

No Known Activations