INDEX

Explanations

words related to political figures or events

oai_token-act-pair · gpt-3.5-turbo

words or terms related to a specific historical or cultural context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

nard

-0.82

rique

-0.75

icas

-0.67

CLASSIFIED

-0.66

ICA

-0.66

 Heard

-0.66

igious

-0.65

icals

-0.65

rican

-0.64

berto

-0.64

POSITIVE LOGITS

pper

1.03

1.01

yi

0.98

yk

0.94

wi

0.87

hler

0.85

yo

0.83

ye

0.83

atchewan

0.82

sie

0.81

Activations Density 0.068%

words related to political figures or events

words or terms related to a specific historical or cultural context

No Comments

No Known Activations