INDEX

Explanations

words related to negative events, disasters, and challenges

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

sonder

-0.60

hever

-0.55

pfe

-0.54

elee

-0.54

 wille

-0.52

 individuel

-0.52

tothe

-0.50

heyd

-0.50

 Neub

-0.50

esss

-0.49

POSITIVE LOGITS

ⓧ

0.66

 tetrach

0.65

 popoli

0.65

 kasama

0.64

ffilm

0.63

 affatto

0.62

 venuto

0.62

<bos>

0.61

 kanya

0.61

 papà

0.59

Activations Density 0.390%

words related to negative events, disasters, and challenges

No Comments

No Known Activations