INDEX

Explanations

references to specific terms or names, potentially related to media productions or locations

oai_token-act-pair · gpt-3.5-turbo

references to storytelling elements and narrative structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

glers

-0.76

rotein

-0.75

erves

-0.66

LECT

-0.65

 acids

-0.64

ophical

-0.64

ophers

-0.63

pocket

-0.62

ensitive

-0.62

entric

-0.62

POSITIVE LOGITS

von

0.83

zan

0.74

pling

0.72

æ©

0.71

Â¯Â¯

0.70

iger

0.70

arov

0.69

orah

0.69

©¶æ¥µ

0.68

 Donetsk

0.66

Activations Density 0.021%

references to specific terms or names, potentially related to media productions or locations

references to storytelling elements and narrative structures

No Comments

No Known Activations