INDEX

Explanations

examples of events or situations that lead to blame being assigned

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

 ?...

-0.95

 fatis

-0.94

 !...

-0.91

 nece

-0.86

fuf

-0.85

 perfon

-0.85

 dichi

-0.82

 Luglio

-0.82

 Dijo

-0.82

 specialmente

-0.81

POSITIVE LOGITS

0.77

0.76

.;

0.68

 because

0.66

but

0.65

0.61

 here

0.61

；

0.59

for

0.59

Activations Density 0.587%

examples of events or situations that lead to blame being assigned

No Comments

No Known Activations