INDEX

Explanations

phrases indicating something as being usual, expected, or frequently occurring

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "common" 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

atever

-0.76

ceans

-0.72

phabet

-0.69

apo

-0.68

usalem

-0.68

oops

-0.68

bern

-0.68

ADRA

-0.66

ignt

-0.66

hens

-0.65

POSITIVE LOGITS

 occurrence

1.19

 occurrences

1.16

 denomin

1.08

alities

1.00

ality

0.99

wealth

0.96

 practice

0.90

 misconception

0.89

 amongst

0.88

places

0.86

Activations Density 0.056%

phrases indicating something as being usual, expected, or frequently occurring

occurrences of the word "common" in various contexts

No Comments

No Known Activations