INDEX

Explanations

phrases related to achieving success or recognition

oai_token-act-pair · gpt-3.5-turbo

terms related to gaining or accumulating something, particularly achievements or support

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

 livest

-0.77

 Shelter

-0.65

 warr

-0.65

 Settlement

-0.64

 Koen

-0.61

 Friendly

-0.60

 friendly

-0.59

Gas

-0.59

 Golf

-0.58

 Loan

-0.58

POSITIVE LOGITS

interstitial

0.97

cffff

0.89

gie

0.80

OGR

0.78

ufact

0.78

uchin

0.78

pload

0.77

igham

0.75

aukee

0.75

anche

0.75

Activations Density 0.053%

phrases related to achieving success or recognition

terms related to gaining or accumulating something, particularly achievements or support

No Comments

No Known Activations