INDEX

Explanations

objects or items related to physical locations or environments

oai_token-act-pair · gpt-3.5-turbo

words related to food, beverages, and nutrition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Reloaded

-0.80

ItemTracker

-0.75

Pwr

-0.75

ADRA

-0.74

 Consortium

-0.71

 Acquisition

-0.70

 Accountability

-0.70

 WATCHED

-0.67

 Blueprint

-0.66

RESULTS

-0.64

POSITIVE LOGITS

erb

0.81

weed

0.80

0.77

umb

0.74

pipe

0.74

aroo

0.73

bars

0.72

mount

0.72

ont

0.72

bol

0.72

Activations Density 0.719%

objects or items related to physical locations or environments

words related to food, beverages, and nutrition

No Comments

No Known Activations