INDEX

Explanations

occurrences of specific numerical values or references within a document

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.01

1:0.01

2:0.03

3:0.13

4:0.07

5:0.16

6:0.02

7:0.04

8:0.19

9:0.06

10:0.06

11:0.15

Negative Logits

ּ

-1.72

gart

-1.60

gars

-1.55

tf

-1.50

ea

-1.44

 disg

-1.41

BILITIES

-1.40

ppings

-1.39

hirt

-1.39

 Gors

-1.37

POSITIVE LOGITS

 campaigner

1.58

 debate

1.56

 contributor

1.55

Experts

1.53

 overhe

1.52

REF

1.51

 campaigning

1.49

 contributors

1.48

REC

1.45

 campaigners

1.45

Activations Density 0.001%

occurrences of specific numerical values or references within a document

No Comments

No Known Activations