INDEX

Explanations

words related to significant events or actions

oai_token-act-pair · gpt-3.5-turbo

words indicating significant actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

rea

-0.67

conn

-0.66

leases

-0.64

rone

-0.63

aly

-0.63

zh

-0.61

lords

-0.60

rel

-0.60

 Newman

-0.60

sw

-0.59

POSITIVE LOGITS

ometimes

1.05

hift

0.95

paces

0.93

omething

0.87

heet

0.86

creen

0.84

ilver

0.83

pace

0.80

hirt

0.79

psey

0.71

Activations Density 0.655%

words related to significant events or actions

words indicating significant actions or events

No Comments

No Known Activations