INDEX

Explanations

words related to specific types of fruits or vegetables

oai_token-act-pair · gpt-3.5-turbo

mentions of pumpkin and squash

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

ugu

-0.80

lasses

-0.80

gencies

-0.77

stood

-0.73

 Cheong

-0.73

Reply

-0.71

nces

-0.71

rators

-0.68

eq

-0.67

 Alam

-0.66

POSITIVE LOGITS

 spice

1.06

fruit

1.00

 pumpkin

0.91

pie

0.88

 juice

0.85

 Spice

0.84

 Pumpkin

0.80

burg

0.80

 pies

0.76

pen

0.75

Activations Density 0.014%

words related to specific types of fruits or vegetables

mentions of pumpkin and squash

No Comments

No Known Activations