INDEX

Explanations

titles or introductory phrases including a specific name, "Kevin Zeese."

oai_token-act-pair · gpt-3.5-turbo

negative contractions of the verb "to be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Reviewer

-0.77

çĦ

-0.71

 behavi

-0.68

soType

-0.67

è»

-0.66

pter

-0.65

LIN

-0.64

itiz

-0.64

 ancest

-0.64

edIn

-0.64

POSITIVE LOGITS

 necessarily

1.18

 bother

1.04

urtles

0.93

ember

0.87

itles

0.86

urtle

0.85

 seem

0.85

 anymore

0.84

aylor

0.83

 exactly

0.81

Activations Density 0.099%

titles or introductory phrases including a specific name, "Kevin Zeese."

negative contractions of the verb "to be."

No Comments

No Known Activations