INDEX

Explanations

phrases that indicate a specific location or context

oai_token-act-pair · gpt-3.5-turbo

references to timeframes and contextual phrases

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

arcity

-0.73

 viol

-0.65

 pian

-0.64

ocre

-0.62

daq

-0.61

 torches

-0.60

 Luigi

-0.59

culosis

-0.59

gery

-0.58

piece

-0.58

POSITIVE LOGITS

 confines

1.58

 bounds

1.27

 boundaries

0.97

 limits

0.93

 borders

0.91

 radius

0.89

 parameters

0.88

ombat

0.86

 realms

0.84

 circles

0.84

Activations Density 0.121%

phrases that indicate a specific location or context

references to timeframes and contextual phrases

No Comments

No Known Activations