INDEX

Explanations

mentions of the name "Ori" followed by a number indicating the strength of the match

oai_token-act-pair · gpt-3.5-turbo

references to the term "ori" 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 @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

esville

-0.75

 afore

-0.71

drivers

-0.71

dream

-0.70

sim

-0.69

etimes

-0.68

ridges

-0.67

session

-0.67

sic

-0.66

checks

-0.65

POSITIVE LOGITS

ori

1.11

Äĩ

0.99

onto

0.93

Äį

0.92

andom

0.92

ya

0.87

atively

0.86

atis

0.86

opsis

0.84

oma

0.84

Activations Density 0.004%

mentions of the name "Ori" followed by a number indicating the strength of the match

references to the term "ori" in various contexts

No Comments

No Known Activations