INDEX

Explanations

actions related to checking or verifying something

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "check" and its variations

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

ufact

-0.71

Ã±a

-0.67

 Hots

-0.66

nown

-0.64

joice

-0.64

SAY

-0.63

nect

-0.62

usable

-0.62

åħī

-0.62

asus

-0.62

POSITIVE LOGITS

mate

1.02

lists

0.91

boxes

0.89

 whether

0.83

out

0.75

 boxes

0.74

points

0.70

ysis

0.70

 balances

0.69

box

0.68

Activations Density 0.033%

actions related to checking or verifying something

occurrences of the word "check" and its variations

No Comments

No Known Activations