INDEX

Explanations

instances of emphasizing importance or specificity

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "that" 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 @ 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

ahime

-0.70

agram

-0.67

izen

-0.66

roth

-0.65

oco

-0.65

vre

-0.65

ardless

-0.64

pling

-0.62

hips

-0.61

ivery

-0.60

POSITIVE LOGITS

 happens

1.40

 happened

1.40

 distinguishes

1.27

 transpired

1.26

 separates

1.21

 occurs

1.13

 bothers

1.12

 prevents

1.06

 relates

1.04

 happen

1.04

Activations Density 0.111%

instances of emphasizing importance or specificity

phrases that include the word "that" in various contexts

No Comments

No Known Activations