INDEX

Explanations

proper nouns associated with recognition or awards

oai_token-act-pair · gpt-3.5-turbo

terms related to inductive categories or classifications in various contexts, particularly in performance and achievement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 feast

-0.70

 truce

-0.70

 reper

-0.68

 constellation

-0.66

 nerv

-0.63

 clen

-0.62

 grip

-0.61

 ende

-0.61

 beet

-0.61

 mockery

-0.60

POSITIVE LOGITS

ers

1.82

ership

1.37

ered

1.31

ors

1.30

erers

1.24

erer

1.23

orship

1.14

ables

1.09

ering

1.08

ename

1.07

Activations Density 0.120%

proper nouns associated with recognition or awards

terms related to inductive categories or classifications in various contexts, particularly in performance and achievement

No Comments

No Known Activations