INDEX

Explanations

indicators of expectation and anticipation in various contexts

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_L7_Hcat_z_lr1.20e-03_l11.10e+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.7.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.11

1:0.05

2:0.02

3:0.14

4:0.06

5:0.13

6:0.03

7:0.02

8:0.08

9:0.26

10:0.01

11:0.01

Negative Logits

arc

-1.80

anon

-1.73

EFF

-1.69

astern

-1.66

Study

-1.62

arest

-1.61

Haunted

-1.59

utterstock

-1.59

dn

-1.58

bleacher

-1.57

POSITIVE LOGITS

soon

2.26

 tomorrow

2.19

pour

1.88

 soon

1.87

ebus

1.78

 anytime

1.73

edo

1.68

 forthcoming

1.66

 future

1.65

 foreseeable

1.64

Activations Density 0.091%

indicators of expectation and anticipation in various contexts

No Comments

No Known Activations