INDEX

Explanations

concepts related to symbolic gestures and expressions of good faith

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

0:0.02

1:0.02

2:0.05

3:0.06

4:0.08

5:0.03

6:0.08

7:0.31

8:0.03

9:0.04

10:0.15

11:0.06

Negative Logits

 diluted

-1.65

Course

-1.57

�

-1.51

 contaminated

-1.50

iencies

-1.49

 researched

-1.46

renheit

-1.45

 Dalton

-1.45

 tracts

-1.42

 Stories

-1.40

POSITIVE LOGITS

 gesture

3.07

 gestures

2.57

 goodbye

1.82

 clasp

1.75

 recognition

1.63

 kindness

1.60

 appreciation

1.60

 greeting

1.59

 wink

1.58

 gratification

1.56

Activations Density 0.001%

concepts related to symbolic gestures and expressions of good faith

No Comments

No Known Activations