INDEX

Explanations

words related to software and technology performance

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 indeb

-0.91

 pecuni

-0.86

 affari

-0.83

 abbra

-0.83

 kompati

-0.76

erk

-0.76

 lende

-0.76

 seduta

-0.74

 socie

-0.74

 soste

-0.73

POSITIVE LOGITS

own

0.68

 latest

0.66

0.64

 masterful

0.61

 penchant

0.60

 infamous

0.59

 insatiable

0.59

 insistence

0.57

 inaugural

0.56

0.55

Activations Density 0.275%

words related to software and technology performance

No Comments

No Known Activations