INDEX

Explanations

mentions of the Republican party or associated political references

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

0:0.08

1:0.07

2:0.09

3:0.09

4:0.08

5:0.07

6:0.09

7:0.07

8:0.07

9:0.07

10:0.07

11:0.08

Negative Logits

ipe

-1.49

 Collection

-1.43

 teasp

-1.42

Sports

-1.41

wl

-1.41

Catal

-1.40

 boutique

-1.40

 Magazine

-1.40

eton

-1.40

 Ward

-1.39

POSITIVE LOGITS

zees

1.89

ModLoader

1.66

 satisf

1.59

laughter

1.48

 ordeal

1.48

 sacrifices

1.47

////////////////

1.47

EngineDebug

1.46

 tink

1.46

 bloodshed

1.45

Activations Density 0.000%

mentions of the Republican party or associated political references

No Comments

No Known Activations