INDEX

Explanations

references to a specific geographic feature - a river

oai_token-act-pair · gpt-3.5-turbo

references to rivers

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

Reloaded

-0.75

Mach

-0.68

erity

-0.67

MFT

-0.62

ACH

-0.62

 Horowitz

-0.61

iaries

-0.61

ernel

-0.61

xus

-0.60

Pie

-0.59

POSITIVE LOGITS

bank

1.03

front

1.02

banks

1.01

 basin

0.94

 river

0.93

 valley

0.92

ine

0.88

 rivers

0.87

 valleys

0.85

 canyon

0.85

Activations Density 0.028%

references to a specific geographic feature - a river

references to rivers

No Comments

No Known Activations