INDEX

Explanations

terms or phrases that act as indicators, often related to measurements or assessments

oai_token-act-pair · gpt-3.5-turbo

terms related to indicators or metrics

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

ernel

-0.83

©¶æ

-0.82

frey

-0.81

lishing

-0.77

awar

-0.74

atche

-0.71

ategory

-0.71

POSE

-0.69

orld

-0.69

cow

-0.69

POSITIVE LOGITS

 indicators

1.08

 markers

0.96

 indicator

0.95

 gauge

0.86

 marker

0.86

 Signs

0.72

 cues

0.72

posts

0.71

icator

0.70

 sidebar

0.67

Activations Density 0.036%

terms or phrases that act as indicators, often related to measurements or assessments

terms related to indicators or metrics

No Comments

No Known Activations