INDEX

Explanations

words related to scams and fraudulent activities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "spam" in various contexts

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

yi

-0.70

 Directions

-0.65

elsius

-0.64

ONSORED

-0.63

 cutoff

-0.61

 kinderg

-0.59

WIND

-0.59

¥ŀ

-0.59

 anticipation

-0.59

 predis

-0.59

POSITIVE LOGITS

nesty

1.22

pling

1.10

sterdam

1.09

ilies

1.08

pering

1.02

borgh

1.01

pler

0.98

oufl

0.98

ilial

0.98

mer

0.97

Activations Density 0.030%

words related to scams and fraudulent activities

occurrences of the word "spam" in various contexts

No Comments

No Known Activations