INDEX

Explanations

mentions of a specific person or entity named "Av"

oai_token-act-pair · gpt-3.5-turbo

references to a specific entity or person named "Av."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

utenant

-0.77

hyde

-0.70

 inaccessible

-0.66

zsche

-0.63

FORMATION

-0.63

ciplinary

-0.61

bered

-0.61

 Barnett

-0.59

 towels

-0.59

 cooker

-0.59

POSITIVE LOGITS

atars

1.12

ril

1.04

ocado

1.03

iator

0.99

iol

0.98

raham

0.98

iew

0.98

iva

0.95

atar

0.94

ille

0.94

Activations Density 0.017%

mentions of a specific person or entity named "Av"

references to a specific entity or person named "Av."

No Comments

No Known Activations