INDEX

Explanations

conjunctions and expressions of conditional or additional clauses

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_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.3.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

2:0.15

3:0.13

4:0.01

5:0.03

6:0.09

7:0.17

8:0.06

9:0.04

10:0.07

11:0.15

Negative Logits

sectional

-0.98

rider

-0.97

VICE

-0.97

quartered

-0.97

nee

-0.92

ghai

-0.91

wife

-0.90

oft

-0.90

incial

-0.90

croft

-0.88

POSITIVE LOGITS

 Meow

1.03

roc

1.01

 toile

1.00

 horm

0.99

 Cheese

0.98

 Metatron

0.94

roo

0.93

oned

0.93

dx

0.91

->

0.91

Activations Density 0.024%

conjunctions and expressions of conditional or additional clauses

No Comments

No Known Activations