INDEX

Explanations

words related to surrounding or enclosing something tightly

oai_token-act-pair · gpt-3.5-turbo

words related to encirclement and various forms of deportation

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

 reckoning

-0.83

 surviving

-0.79

 trump

-0.65

 nerv

-0.62

 scraps

-0.62

 spine

-0.61

 scales

-0.61

yip

-0.59

 Ancients

-0.59

 dearly

-0.59

POSITIVE LOGITS

mental

1.08

ment

0.95

ciating

0.90

agement

0.89

ãĤ¼ãĤ¦ãĤ¹

0.84

enced

0.84

ados

0.83

ption

0.83

cheon

0.81

achment

0.81

Activations Density 0.071%

words related to surrounding or enclosing something tightly

words related to encirclement and various forms of deportation

No Comments

No Known Activations