INDEX

Explanations

phrases related to boundaries or limits being crossed

oai_token-act-pair · gpt-3.5-turbo

references to boundaries or thresholds being crossed

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 followed

-0.81

FORE

-0.79

 alone

-0.78

MRI

-0.77

 apart

-0.74

 aloud

-0.73

 alike

-0.73

 anew

-0.72

CLA

-0.72

DAY

-0.70

POSITIVE LOGITS

 threshold

1.42

 brink

1.17

 thresholds

1.17

 boundaries

1.10

 limit

1.09

 limits

1.05

 bounds

1.04

 fray

1.01

 mark

1.00

 requisite

0.98

Activations Density 0.154%

phrases related to boundaries or limits being crossed

references to boundaries or thresholds being crossed

No Comments

No Known Activations