INDEX

Explanations

instances where actions or plans are being suggested or executed

oai_token-act-pair · gpt-3.5-turbo

phrases related to the act of implementing policies or strategies

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

pool

-0.70

si

-0.67

bery

-0.66

 Tire

-0.65

mpeg

-0.63

rooms

-0.61

Sul

-0.60

 sands

-0.60

weed

-0.57

ping

-0.57

POSITIVE LOGITS

urally

0.82

uate

0.80

ually

0.79

ulate

0.78

eering

0.74

 implementing

0.73

 implementation

0.73

 implemented

0.72

matically

0.70

mental

0.70

Activations Density 0.020%

instances where actions or plans are being suggested or executed

phrases related to the act of implementing policies or strategies

No Comments

No Known Activations