INDEX

Explanations

the word "Never" followed by a verb

oai_token-act-pair · gpt-3.5-turbo

the repeated phrase "Never" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

eers

-0.69

redients

-0.67

iency

-0.65

uliffe

-0.64

?????-

-0.61

AQ

-0.60

jug

-0.60

ipation

-0.59

 Jindal

-0.58

adobe

-0.57

POSITIVE LOGITS

theless

1.78

winter

1.01

mind

0.86

 Alone

0.79

 ceases

0.77

entimes

0.75

 Forget

0.72

 EVER

0.71

epad

0.70

BuyableInstoreAndOnline

0.70

Activations Density 0.036%

the word "Never" followed by a verb

the repeated phrase "Never" in various contexts

No Comments

No Known Activations