INDEX

Explanations

adjectives and verbs associated with actions or decisions

oai_token-act-pair · gpt-3.5-turbo

assertions and conditions about actions and outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Tracks

-0.51

 Info

-0.50

oS

-0.49

 league

-0.47

arsity

-0.47

 duties

-0.46

 Instr

-0.45

 Their

-0.45

 radios

-0.45

 tatt

-0.44

POSITIVE LOGITS

.ï¿½

0.73

.$

0.73

.</

0.72

.''

0.72

[/

0.71

EStream

0.70

 unthinkable

0.68

.<

0.67

.(

0.66

EStreamFrame

0.65

Activations Density 0.858%

adjectives and verbs associated with actions or decisions

assertions and conditions about actions and outcomes

No Comments

No Known Activations