INDEX

Explanations

words that indicate actions or states of being

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 indebted

-0.16

shal

-0.15

Ð°ÑĢÐ°Ð¼

-0.15

paged

-0.14

 Ð½Ð°Ð²

-0.14

hl

-0.14

Ð´Ð°Ð»

-0.14

 ÄĲÃ´ng

-0.14

exampleInput

-0.14

fac

-0.14

POSITIVE LOGITS

ibi

0.14

Cul

0.14

 nackte

0.14

 Samp

0.14

.uni

0.14

atoi

0.14

 capped

0.14

ovit

0.14

enda

0.13

*(*

0.13

Activations Density 0.002%

words that indicate actions or states of being

No Comments

No Known Activations