INDEX

Explanations

No Explanations Found

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

0:0.07

1:0.08

2:0.08

3:0.07

4:0.08

5:0.08

6:0.08

7:0.08

8:0.08

9:0.08

10:0.09

11:0.09

Negative Logits

 winners

-2.43

 hawk

-2.17

 Oscars

-2.06

 Peaks

-2.00

 Noon

-2.00

 precedence

-1.96

abad

-1.95

 kings

-1.93

��

-1.93

 distinctions

-1.85

POSITIVE LOGITS

acl

2.23

DEBUG

2.12

inder

2.09

owered

2.09

repl

2.04

SPONSORED

2.04

milo

2.01

involved

1.98

skinned

1.93

checking

1.88

Activations Density 0.000%

No Known Activations

This feature has no known activations.