INDEX

Explanations

text related to predictions and consequences

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

 kasa

-0.76

Okt

-0.71

 fers

-0.70

 lapto

-0.69

 lele

-0.69

 traktor

-0.69

 karton

-0.66

kac

-0.65

laf

-0.65

 kade

-0.64

POSITIVE LOGITS

 predictions

1.22

 predicting

1.19

 prediction

1.17

 predict

1.10

 Predictions

1.07

 predictive

1.07

 predicts

1.04

 Prediction

1.02

 predicted

1.02

 forecasting

1.01

Activations Density 0.457%

text related to predictions and consequences

No Comments

No Known Activations