INDEX

Explanations

phrases that indicate omitted information or facts

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

2:0.08

3:0.05

4:0.14

5:0.02

6:0.05

7:0.41

8:0.04

9:0.03

10:0.06

11:0.06

Negative Logits

anic

-1.66

orses

-1.65

rha

-1.60

lav

-1.60

wagen

-1.55

ivot

-1.53

yrinth

-1.53

oros

-1.51

rg

-1.51

oled

-1.50

POSITIVE LOGITS

 altogether

2.01

 anymore

1.80

 distinctions

1.64

 jokes

1.62

 incidentally

1.54

 redund

1.47

 because

1.44

 comparisons

1.44

 Doodle

1.42

 comparison

1.41

Activations Density 0.004%

phrases that indicate omitted information or facts

No Comments

No Known Activations