INDEX

Explanations

words related to strong attributes or actions

oai_token-act-pair · gpt-3.5-turbo

terms related to concepts of stability, support, and positive attributes in various 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

 peas

-0.76

 charts

-0.65

 usual

-0.65

 torches

-0.63

 calcul

-0.63

 needles

-0.61

rums

-0.61

 flares

-0.61

 diagrams

-0.61

ighters

-0.60

POSITIVE LOGITS

 consisting

0.78

(>

0.74

cko

0.74

,,,,

0.73

 compared

0.72

pour

0.72

kick

0.71

enance

0.71

 towards

0.70

Initialized

0.70

Activations Density 0.362%

words related to strong attributes or actions

terms related to concepts of stability, support, and positive attributes in various contexts

No Comments

No Known Activations