INDEX

Explanations

terms related to "secondary" functions, effects, or characteristics

oai_token-act-pair · gpt-3.5-turbo

references to secondary effects or secondary characteristics 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ebus

-0.83

Roberts

-0.76

âĸĪâĸĪ

-0.69

===

-0.69

efully

-0.66

 Archdemon

-0.66

âĸĪâĸĪâĸĪâĸĪ

-0.65

âķĲâķĲ

-0.65

ERG

-0.64

Extreme

-0.63

POSITIVE LOGITS

secondary

0.99

 secondary

0.95

 mosqu

0.77

 Secondary

0.75

arily

0.74

pole

0.73

 tert

0.69

 mater

0.68

 primary

0.68

nces

0.68

Activations Density 0.009%

terms related to "secondary" functions, effects, or characteristics

references to secondary effects or secondary characteristics in various contexts

No Comments

No Known Activations