INDEX

Explanations

occurrences of the word "cases"

oai_token-act-pair · gpt-3.5-turbo

phrases indicating examples or instances of situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

asus

-0.73

roe

-0.71

liction

-0.68

 livest

-0.68

 slic

-0.67

CLASSIFIED

-0.66

obil

-0.65

BN

-0.64

redit

-0.61

 streng

-0.60

POSITIVE LOGITS

cases

0.92

hips

0.84

hooting

0.83

 cases

0.82

cale

0.81

 involving

0.81

manship

0.81

hip

0.78

afety

0.75

mith

0.75

Activations Density 0.028%

occurrences of the word "cases"

phrases indicating examples or instances of situations

No Comments

No Known Activations