INDEX

Explanations

phrases indicating transformation or combination

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

0:0.02

1:0.02

2:0.07

3:0.08

4:0.14

5:0.03

6:0.03

7:0.40

8:0.03

9:0.04

10:0.04

11:0.04

Negative Logits

articles

-1.69

missing

-1.61

Recommend

-1.60

listed

-1.57

flix

-1.53

lost

-1.50

Tickets

-1.48

-1.47

bugs

-1.47

Missing

-1.46

POSITIVE LOGITS

 plur

1.66

 seamlessly

1.65

 INTO

1.64

 into

1.55

 symb

1.54

 rapport

1.53

 Into

1.51

 cohesive

1.51

 normalized

1.50

 continuity

1.49

Activations Density 0.001%

phrases indicating transformation or combination

No Comments

No Known Activations