INDEX

Explanations

references to or mentions of Allied forces

oai_token-act-pair · gpt-3.5-turbo

terms related to military alliances and forces

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

estamp

-0.69

Cu

-0.68

mys

-0.67

ema

-0.66

bed

-0.65

riet

-0.64

 grammar

-0.64

 github

-0.63

uc

-0.63

 rash

-0.63

POSITIVE LOGITS

 Allied

3.93

 Allies

2.61

 allied

2.08

 Axis

1.86

 postwar

1.22

 Ottoman

1.19

 Ally

1.18

RAF

1.12

 Eisenhower

1.11

 Alliance

1.10

Activations Density 0.028%

references to or mentions of Allied forces

terms related to military alliances and forces

No Comments

No Known Activations