INDEX

Explanations

words containing the string "ams"

oai_token-act-pair · gpt-3.5-turbo

the presence of the term "ams" 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 resemb

-0.65

Charge

-0.65

 rupture

-0.64

stood

-0.63

 escalation

-0.62

 reversal

-0.62

 draw

-0.62

 acknowledgment

-0.62

 partisans

-0.61

 referral

-0.60

POSITIVE LOGITS

arah

1.08

sterdam

1.07

ams

1.03

borgh

1.01

mith

1.00

olver

0.98

alam

0.97

onde

0.96

heet

0.95

ilton

0.94

Activations Density 0.008%

words containing the string "ams"

the presence of the term "ams" in various contexts

No Comments

No Known Activations