INDEX

Explanations

words related to transportation and physical objects

oai_token-act-pair · gpt-3.5-turbo

objects and items often associated with logistical or security contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

âĸ¬

-0.91

 experien

-0.82

ãĥį

-0.76

history

-0.72

Rate

-0.69

ãĥ´

-0.68

 practition

-0.68

ãĤ´ãĥ³

-0.67

AAAAAAAA

-0.67

Relations

-0.63

POSITIVE LOGITS

poons

1.30

hips

1.22

mith

1.10

creen

1.02

leeve

0.99

pring

0.96

uits

0.96

pots

0.92

poon

0.91

 stacked

0.89

Activations Density 0.358%

words related to transportation and physical objects

objects and items often associated with logistical or security contexts

No Comments

No Known Activations