INDEX

Explanations

the word "that"

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

0:0.08

1:0.05

2:0.07

3:0.08

4:0.08

5:0.07

6:0.08

7:0.09

8:0.09

9:0.07

10:0.09

11:0.08

Negative Logits

issance

-2.02

��

-1.86

rica

-1.74

CVE

-1.71

��

-1.69

��

-1.69

 Helm

-1.67

})

-1.64

 Caucasus

-1.59

Secondly

-1.57

POSITIVE LOGITS

 exclaim

1.73

offer

1.65

 keynote

1.62

 reject

1.62

 cance

1.62

 cancel

1.61

 approve

1.59

 phone

1.57

 filings

1.56

 unsuccessfully

1.55

Activations Density 0.000%

the word "that"

No Comments

No Known Activations