INDEX

Explanations

names or mentions of individuals or organizations

oai_token-act-pair · gpt-3.5-turbo

specific letters, predominantly 'c' and 'd', within the text

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

å§«

-0.75

LEASE

-0.72

SUP

-0.68

é¾

-0.67

BILITY

-0.65

è»

-0.65

IGHTS

-0.65

 bells

-0.65

ebted

-0.64

åħī

-0.64

POSITIVE LOGITS

uta

0.86

ama

0.84

ara

0.82

ula

0.81

oslav

0.79

ana

0.79

ka

0.76

ul

0.75

ija

0.75

Si

0.74

Activations Density 0.152%

names or mentions of individuals or organizations

specific letters, predominantly 'c' and 'd', within the text

No Comments

No Known Activations