INDEX

Explanations

words related to providing support or evidence for a claim or action

oai_token-act-pair · gpt-3.5-turbo

words related to support or endorsement

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

ities

-0.80

urious

-0.74

 ILCS

-0.73

illin

-0.70

 Hots

-0.68

hester

-0.68

alde

-0.67

istics

-0.65

anny

-0.65

thora

-0.64

POSITIVE LOGITS

backed

0.79

abies

0.79

track

0.77

GROUND

0.77

 backing

0.75

drive

0.74

steen

0.72

raise

0.71

swing

0.71

drops

0.69

Activations Density 0.024%

words related to providing support or evidence for a claim or action

words related to support or endorsement

No Comments

No Known Activations