INDEX

Explanations

phrases expressing equivalence or comparisons

oai_token-act-pair · gpt-3.5-turbo

comparisons indicating equivalence 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 @ 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

stal

-0.82

hra

-0.77

spe

-0.76

stra

-0.74

oard

-0.74

 Bomb

-0.68

omen

-0.67

bean

-0.67

 Roads

-0.65

 Mush

-0.65

POSITIVE LOGITS

ivalent

0.84

lihood

0.83

isons

0.82

imately

0.80

 amounts

0.77

terday

0.77

icut

0.74

aminer

0.72

 equivalent

0.72

oreal

0.72

Activations Density 0.016%

phrases expressing equivalence or comparisons

comparisons indicating equivalence in various contexts

No Comments

No Known Activations