INDEX

Explanations

phrases related to quotes or statements being said

oai_token-act-pair · gpt-3.5-turbo

periods at the end of statements

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

 skirm

-0.74

ensibly

-0.68

 spoiled

-0.67

 assassin

-0.66

 silly

-0.66

 slightest

-0.65

toy

-0.64

 midrange

-0.64

 unlucky

-0.64

 transact

-0.64

POSITIVE LOGITS

Refer

0.96

↵↵

0.89

Elsewhere

0.85

[+

0.84

 Another

0.83

Earlier

0.82

org

0.82

 Accessed

0.81

 Meanwhile

0.81

<|endoftext|>

0.80

Activations Density 0.210%

phrases related to quotes or statements being said

periods at the end of statements

No Comments

No Known Activations