INDEX

Explanations

data and figures related to box office performance

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

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.03

1:0.01

2:0.06

3:0.06

4:0.05

5:0.03

6:0.04

7:0.08

8:0.04

9:0.06

10:0.40

11:0.08

Negative Logits

chan

-2.23

 pronoun

-1.89

 Emacs

-1.88

 noun

-1.82

 deprecated

-1.81

 pronouns

-1.75

 hairst

-1.74

verb

-1.74

 symbol

-1.74

Config

-1.72

POSITIVE LOGITS

 averages

1.98

 Brisbane

1.95

 peaks

1.84

 Trends

1.78

 Estimated

1.78

AUD

1.75

 Overall

1.74

Tri

1.72

 Prediction

1.69

 Average

1.68

Activations Density 0.005%

data and figures related to box office performance

No Comments

No Known Activations